1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
8 and 3.4.4 - once the server is enabled to respond to an OCSP request
9 it does even when not requested, resulting in a stapling non-aware
10 client dropping the TLS connection.
12 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
13 support variable-length bit vectors. No functional change.
15 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
16 The I= interface field on outgoing lines is now after the H= remote
17 host field, same as incoming lines. There is a separate
18 outgoing_interface log selector which allows you to disable the
21 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
22 If not running log_selector +smtp_connection the mainlog would be held
23 open indefinitely after a "too many connections" event, including to a
24 deleted file after a log rotate. Leave the per net connection logging
25 leaving it open for efficiency as that will be quickly detected by the
26 check on the next write.
28 HS/01 Bug 1671: Fix post transport crash.
29 Processing the wait-<transport> messages could crash the delivery
30 process if the message IDs didn't exist for some reason. When
31 using 'split_spool_directory=yes' the construction of the spool
32 file name failed already, exposing the same netto behaviour.
34 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
35 mime_regex ACL conditions.
37 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
38 to DSN fail messages (bounces): remote IP, remote greeting, remote response
39 to HELO, local diagnostic string.
41 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
42 log line to debug. Even when configured with a tls authenticator many
43 client connections are expected to not authenticate in this way, so
44 an authenticate fail is not an error.
46 HS/02 Add the Exim version string to the process info. This way exiwhat
47 gives some more detail about the running daemon.
49 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
50 matter for fast-change records such as DNSBLs.
52 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
53 retry record, even if constant. There may be multiple transports with
54 different interface settings and the retry behaviour needs to be kept
57 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
59 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
61 JH/10 Bug 840: fix log_defer_output option of pipe transport
63 JH/11 Bug 830: use same host for all RCPTS of a message, even under
64 hosts_randomize. This matters a lot when combined with mua_wrapper.
66 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
67 ${quote_pgsql:<string>} operator.
69 JH/13 Bug 1708: avoid misaligned access in cached lookup.
71 JH/14 Change header file name for freeradius-client. Relevant if compiling
72 with Radius support; from the Gentoo tree and checked under Fedora.
74 JH/15 Bug 1712: Introduce $prdr_requested flag variable
76 JH/16 Bug 1714: Permit an empty string as expansion result for transport
77 option transport_filter, meaning no filtering.
79 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
81 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
82 defaults to "*" (all hosts). The variable is now available when not built
83 with TLS, default unset, mainly to enable keeping the testuite sane.
84 If a server certificate is not supplied (via tls_certificate) an error is
85 logged, and clients will find TLS connections fail on startup. Presumably
86 they will retry in-clear.
87 Packagers of Exim are strongly encouraged to create a server certificate
90 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
91 with the $config_file variable.
93 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
94 in transport context, after the attempt, and per-recipient. The latter type
95 is per host attempted. The event data is the error message, and the errno
96 information encodes the lookup type (A vs. MX) used for the (first) host,
97 and the trailing two digits of the smtp 4xx reponse.
99 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
100 to write to mainlog (or rejectlog, paniclog) in the window between file
101 creation and permissions/ownership being changed. Particularly affects
102 installations where exicyclog is run as root, rather than exim user;
103 result is that the running daemon panics and dies.
105 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
107 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
108 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
109 "pri" and "weight". Note that the previous implicit priority given by the
110 list order is no longer honoured.
112 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
115 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
116 by defining SUPPORT_SOCKS.
118 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
119 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
120 is renamed to hosts_proxy.
125 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
128 JH/02 The smtp transport option "multi_domain" is now expanded.
130 JH/03 The smtp transport now requests PRDR by default, if the server offers
133 JH/04 Certificate name checking on server certificates, when exim is a client,
134 is now done by default. The transport option tls_verify_cert_hostnames
135 can be used to disable this per-host. The build option
136 EXPERIMENTAL_CERTNAMES is withdrawn.
138 JH/05 The value of the tls_verify_certificates smtp transport and main options
139 default to the word "system" to access the system default CA bundle.
140 For GnuTLS, only version 3.0.20 or later.
142 JH/06 Verification of the server certificate for a TLS connection is now tried
143 (but not required) by default. The verification status is now logged by
144 default, for both outbound TLS and client-certificate supplying inbound
147 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
150 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
151 Status Notification (bounce) messages are now MIME format per RFC 3464.
152 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
153 under the control of the dsn_advertise_hosts option, and routers may
154 have a dsn_lasthop option.
156 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
157 default, modifiable by a malware= option. The list separator for
158 the options can now be changed in the usual way. Bug 68.
160 JH/10 The smtp_receive_timeout main option is now expanded before use.
162 JH/11 The incoming_interface log option now also enables logging of the
163 local interface on delivery outgoing connections.
165 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
166 if the interface and destination host and port all match.
168 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
171 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
172 Patch from Andrew Lewis.
174 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
175 now supports optional time-restrictions, weighting, and priority
176 modifiers per server. Patch originally by <rommer@active.by>.
178 JH/16 The spamd_address main option now supports a mixed list of local
179 and remote servers. Remote servers can be IPv6 addresses, and
180 specify a port-range.
182 JH/17 Bug 68: The spamd_address main option now supports an optional
183 timeout value per server.
185 JH/18 Bug 1581: Router and transport options headers_add/remove can
186 now have the list separator specified.
188 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
191 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
194 JH/21 Support for the A6 type of dns record is withdrawn.
196 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
197 rather than the verbs used.
199 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
200 from 255 to 1024 chars.
202 JH/24 Verification callouts now attempt to use TLS by default.
204 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
205 are generic router options now. The defaults didn't change.
207 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
208 Original patch from Alexander Shikoff, worked over by JH.
210 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
211 files if ZCAT_COMMAND is not executable.
213 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
215 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
217 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
218 Normally benign, it bites when the pair was led to by a CNAME;
219 modern usage is to not canoicalize the domain to a CNAME target
220 (and we were inconsistent anyway for A-only vs AAAA+A).
222 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
224 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
225 when evaluating $sender_host_dnssec.
227 JH/31 Check the HELO verification lookup for DNSSEC, adding new
228 $sender_helo_dnssec variable.
230 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
232 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
234 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
236 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
237 documented as working, but never had. Support all but $spam_report.
239 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
240 added for tls authenticator.
245 TL/01 When running the test suite, the README says that variables such as
246 no_msglog_check are global and can be placed anywhere in a specific
247 test's script, however it was observed that placement needed to be near
248 the beginning for it to behave that way. Changed the runtest perl
249 script to read through the entire script once to detect and set these
250 variables, reset to the beginning of the script, and then run through
251 the script parsing/test process like normal.
253 TL/02 The BSD's have an arc4random API. One of the functions to induce
254 adding randomness was arc4random_stir(), but it has been removed in
255 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
256 function when detected.
258 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
259 cause callback expansion.
261 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
262 syntax errors in an expansion can be treated as a string instead of
263 logging or causing an error, due to the internal use of bool_lax
264 instead of bool when processing it.
266 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
267 server certificates when making smtp deliveries.
269 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
271 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
273 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
275 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
276 Merged patch from Sebastian Wiedenroth.
278 JH/05 Fix results-pipe from transport process. Several recipients, combined
279 with certificate use, exposed issues where response data items split
280 over buffer boundaries were not parsed properly. This eventually
281 resulted in duplicates being sent. This issue only became common enough
282 to notice due to the introduction of conection certificate information,
283 the item size being so much larger. Found and fixed by Wolfgang Breyha.
285 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
286 size buffer was used, resulting in syntax errors when an expansion
289 JH/07 Add support for directories of certificates when compiled with a GnuTLS
290 version 3.3.6 or later.
292 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
293 is EXPERIMENTAL_EVENT, the main-configuration and transport options
294 both become "event_action", the variables become $event_name, $event_data
295 and $event_defer_errno. There is a new variable $verify_mode, usable in
296 routers, transports and related events. The tls:cert event is now also
297 raised for inbound connections, if the main configuration event_action
300 TL/06 In test suite, disable OCSP for old versions of openssl which contained
301 early OCSP support, but no stapling (appears to be less than 1.0.0).
303 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
304 server certificate names available under the smtp transport option
305 "tls_verify_cert_hostname" now do not permit multi-component wildcard
308 JH/10 Time-related extraction expansions from certificates now use the main
309 option "timezone" setting for output formatting, and are consistent
310 between OpenSSL and GnuTLS compilations. Bug 1541.
312 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
313 encoded parameter in the incoming message. Bug 1558.
315 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
316 include certificate info, eximon was claiming there were spoolfile
319 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
321 JH/14 Log delivery-related information more consistently, using the sequence
322 "H=<name> [<ip>]" wherever possible.
324 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
325 are problematic for Debian distribution, omit them from the release
328 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
330 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
333 JH/17 Fix a null-indirection in certextract expansions when a nondefault
334 output list separator was used.
339 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
340 checkers that were complaining about end of non-void function with no
343 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
344 This was a regression intruduced in 4.83 by another bugfix.
346 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
348 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
349 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
355 TF/01 Correctly close the server side of TLS when forking for delivery.
357 When a message was received over SMTP with TLS, Exim failed to clear up
358 the incoming connection properly after forking off the child process to
359 deliver the message. In some situations the subsequent outgoing
360 delivery connection happened to have the same fd number as the incoming
361 connection previously had. Exim would try to use TLS and fail, logging
362 a "Bad file descriptor" error.
364 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
365 utilities have not been installed.
367 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
368 temporary space as the ACL may create new global variables.
370 TL/01 LDAP support uses per connection or global context settings, depending
371 upon the detected version of the libraries at build time.
373 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
374 to extract and use the src ip:port in logging and expansions as if it
375 were a direct connection from the outside internet. PPv2 support was
376 updated based on HAProxy spec change in May 2014.
378 JH/02 Add ${listextract {number}{list}{success}{fail}}.
380 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
381 Properly escape header and check for NULL return.
383 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
386 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
388 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
389 characters in header names, implemented as a verify condition.
390 Contributed by Michael Fischer v. Mollard.
392 TL/05 Rename SPF condition results err_perm and err_temp to standardized
393 results permerror and temperror. Previous values are deprecated but
394 still accepted. In a future release, err_perm and err_temp will be
395 completely removed, which will be a backward incompatibility if the
396 ACL tests for either of these two old results. Patch contributed by
397 user bes-internal on the mailing list.
399 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
401 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
402 selectors, in both main and reject logs.
404 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
407 JH/07 Add malware type "sock" for talking to simple daemon.
409 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
411 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
412 routers/transports under cutthrough routing.
414 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
415 numbers. Touch up "bool" conditional to keep the same definition.
417 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
419 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
421 JH/12 Expand items in router/transport headers_add or headers_remove lists
422 individually rather than the list as a whole. Bug 1452.
424 Required for reasonable handling of multiple headers_ options when
425 they may be empty; requires that headers_remove items with embedded
426 colons must have them doubled (or the list-separator changed).
428 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
429 view the policy declared in the DMARC record. Currently, $dmarc_status
430 is a combined value of both the record presence and the result of the
433 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
435 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
436 dnslookup router and the smtp transport (applying to the forward
439 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
440 of ldap servers used for a specific lookup. Patch provided by Heiko
443 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
444 New variable $lookup_dnssec_authenticated for observability.
446 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
447 Patch submitted by Lars Timman.
449 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
451 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
452 Requires trusted mode and valid format message id, aborts otherwise.
453 Patch contributed by Heiko Schlichting.
455 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
456 certextract with support for various fields. Bug 1358.
458 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
459 is requested by default, modifiable by smtp transport option
462 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
463 operate on certificate variables to give certificate fingerprints
464 Also new ${sha256:cert_variable}.
466 JH/23 The PRDR feature is moved from being Experimental into the mainline.
468 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
469 Christian Aistleitner.
471 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
473 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
474 file. Patch from Wolfgang Breyha.
476 JH/25 Expand the coverage of the delivery $host and $host_address to
477 client authenticators run in verify callout. Bug 1476.
479 JH/26 Port service names are now accepted for tls_on_connect_ports, to
480 align with daemon_smtp_ports. Bug 72.
482 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
483 support and error reporting did not work properly.
485 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
486 and is readable. Patch from Andrew Colin Kissa.
488 TL/14 Enhance documentation of ${run expansion and how it parses the
489 commandline after expansion, particularly in the case when an
490 unquoted variable expansion results in an empty value.
492 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
494 PP/02 Fix internal collision of T_APL on systems which support RFC3123
495 by renaming away from it. Addresses GH issue 15, reported by
498 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
500 TL/15 SECURITY: prevent double expansion in math comparison functions
501 (can expand unsanitized data). Not remotely exploitable.
508 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
510 PP/02 Make -n do something, by making it not do something.
511 When combined with -bP, the name of an option is not output.
513 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
516 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
517 $sender_host_name and config options to manage this, and basic check
520 PP/05 DSCP support for outbound connections and control modifier for inbound.
522 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
523 (Only plugin which currently uses this is kerberos4, which nobody should
524 be using, but we should make it available and other future plugins might
525 conceivably use it, even though it would break NAT; stuff *should* be
526 using channel bindings instead).
528 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
529 name; added for Sendmail compatibility; requires admin caller.
530 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
531 just ignore it); requires trusted caller.
532 Also parse but ignore: -Ac -Am -X<logfile>
535 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
537 TL/02 Add +smtp_confirmation as a default logging option.
539 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
540 Patch by Magnus Holmgren from 2007-02-20.
542 TL/04 Bugzilla 1281 - Spec typo.
543 Bugzilla 1283 - Spec typo.
544 Bugzilla 1290 - Spec grammar fixes.
546 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
548 TL/06 Add Experimental DMARC support using libopendmarc libraries.
550 TL/07 Fix an out of order global option causing a segfault. Reported to dev
551 mailing list by by Dmitry Isaikin.
553 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
555 JH/02 Support "G" suffix to numbers in ${if comparisons.
557 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
559 NM/01 Bugzilla 1197 - Spec typo
560 Bugzilla 1196 - Spec examples corrections
562 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
564 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
565 gnutls_enable_pkcs11, but renamed to more accurately indicate its
568 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
569 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
571 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
572 "acl {{name}{arg}...}", and optional args on acl condition
575 JH/05 Permit multiple router/transport headers_add/remove lines.
577 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
579 JH/07 Avoid using a waiting database for a single-message-only transport.
580 Performance patch from Paul Fisher. Bugzilla 1262.
582 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
585 JH/09 Add $headers_added variable, with content from use of ACL modifier
586 add_header (but not yet added to the message). Bugzilla 199.
588 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
589 Pulled from Bugzilla 817 by Wolfgang Breyha.
591 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
593 (nb: this is the same fix as in Exim 4.80.1)
595 JH/11 Add A= logging on delivery lines, and a client_set_id option on
598 JH/12 Add optional authenticated_sender logging to A= and a log_selector
601 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
603 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
604 advertise SMTP AUTH mechanism to us, instead of a generic
605 protocol violation error. Also, make Exim more robust to bad
606 data from the Dovecot auth socket.
608 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
610 When a queue runner is handling a message, Exim first routes the
611 recipient addresses, during which it prunes them based on the retry
612 hints database. After that it attempts to deliver the message to
613 any remaining recipients. It then updates the hints database using
616 So if a recipient address works intermittently, it can get repeatedly
617 deferred at routing time. The retry hints record remains fresh so the
618 address never reaches the final cutoff time.
620 This is a fairly common occurrence when a user is bumping up against
621 their storage quota. Exim had some logic in its local delivery code
622 to deal with this. However it did not apply to per-recipient defers
623 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
625 This change adds a proper retry rule check during routing so that the
626 final cutoff time is checked against the message's age. We only do
627 this check if there is an address retry record and there is not a
628 domain retry record; this implies that previous attempts to handle
629 the address had the retry_use_local_parts option turned on. We use
630 this as an approximation for the destination being like a local
631 delivery, as in LMTP.
633 I suspect this new check makes the old local delivery cutoff check
634 redundant, but I have not verified this so I left the code in place.
636 TF/02 Correct gecos expansion when From: is a prefix of the username.
638 Test 0254 submits a message to Exim with the header
642 When I ran the test suite under the user fanf2, Exim expanded
643 the header to contain my full name, whereas it should have added
644 a Resent-Sender: header. It erroneously treats any prefix of the
645 username as equal to the username.
647 This change corrects that bug.
649 GF/01 DCC debug and logging tidyup
650 Error conditions log to paniclog rather than rejectlog.
651 Debug lines prefixed by "DCC: " to remove any ambiguity.
653 TF/03 Avoid unnecessary rebuilds of lookup-related code.
655 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
656 Bug spotted by Jeremy Harris; was flawed since initial commit.
657 Would have resulted in OCSP responses post-SNI triggering an Exim
658 NULL dereference and crash.
660 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
662 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
663 Bug detection, analysis and fix by Samuel Thibault.
664 Bugzilla 1331, Debian bug #698092.
666 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
668 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
669 Server implementation by Todd Lyons, client by JH.
670 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
671 config variable "prdr_enable" controls whether the server
672 advertises the facility. If the client requests PRDR a new
673 acl_data_smtp_prdr ACL is called once for each recipient, after
674 the body content is received and before the acl_smtp_data ACL.
675 The client is controlled by bolth of: a hosts_try_prdr option
676 on the smtp transport, and the server advertisement.
677 Default client logging of deliveries and rejections involving
678 PRDR are flagged with the string "PRDR".
680 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
681 fclose(). Diagnosis by Todd Lyons.
683 PP/17 Update configure.default to handle IPv6 localhost better.
684 Patch by Alain Williams (plus minor tweaks).
687 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
688 This is now consistent with GnuTLS, and is now documented: the
689 previous undocumented portable approach to treating the option as
690 unset was to force an expansion failure. That still works, and
691 an empty string is now equivalent.
693 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
694 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
695 not performing validation itself.
697 PP/20 Added force_command boolean option to pipe transport.
698 Patch from Nick Koston, of cPanel Inc.
700 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
703 TF/04 Added udpsend ACL modifer and hexquote expansion operator
705 PP/21 Fix eximon continuous updating with timestamped log-files.
706 Broken in a format-string cleanup in 4.80, missed when I repaired the
707 other false fix of the same issue.
708 Report and fix from Heiko Schlichting.
711 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
712 Report from Prashanth Katuri.
714 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
715 It's SecureTransport, so affects any MacOS clients which use the
716 system-integrated TLS libraries, including email clients.
718 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
719 using a MIME ACL for non-SMTP local injection.
720 Report and assistance in diagnosis by Warren Baker.
722 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
724 JH/16 Fix comparisons for 64b. Bugzilla 1385.
726 TL/09 Add expansion variable $authenticated_fail_id to keep track of
727 last id that failed so it may be referenced in subsequent ACL's.
729 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
732 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
733 ldap library initialization, allowing self-signed CA's to be
734 used. Also properly sets require_cert option later in code by
735 using NULL (global ldap config) instead of ldap handle (per
736 session). Bug diagnosis and testing by alxgomz.
738 TL/12 Enhanced documentation in the ratelimit.pl script provided in
739 the src/util/ subdirectory.
741 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
742 renamed to Transport Post Delivery Action by Jeremy Harris, as
745 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
746 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
747 redis_servers = needs to be configured which will be used by the redis
748 lookup. Patch from Warren Baker, of The Packet Hub.
750 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
752 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
753 hostname or reverse DNS when processing a host list. Used suggestions
754 from multiple comments on this bug.
756 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
758 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
759 Missed a few lines, added it to make the runtest require no keyboard
762 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
763 contains upper case chars. Make router use caseful_local_part.
765 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
766 support when GnuTLS has been built with p11-kit.
772 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
774 This, or similar/improved, will also be change PP/11 of 4.82.
780 PP/01 Handle short writes when writing local log-files.
781 In practice, only affects FreeBSD (8 onwards).
782 Bugzilla 1053, with thanks to Dmitry Isaikin.
784 NM/01 Bugzilla 949 - Documentation tweak
786 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
789 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
791 PP/02 Implemented gsasl authenticator.
793 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
795 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
796 `pkg-config foo` for cflags/libs.
798 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
799 with rest of GSASL and with heimdal_gssapi.
801 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
802 `pkg-config foo` for cflags/libs for the TLS implementation.
804 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
805 properties get this fed in as external SSF. A number of robustness
806 and debugging improvements to the cyrus_sasl authenticator.
808 PP/08 cyrus_sasl server now expands the server_realm option.
810 PP/09 Bugzilla 1214 - Log authentication information in reject log.
811 Patch by Jeremy Harris.
813 PP/10 Added dbmjz lookup type.
815 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
817 PP/12 MAIL args handles TAB as well as SP, for better interop with
818 non-compliant senders.
819 Analysis and variant patch by Todd Lyons.
821 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
822 Bug report from Lars Müller <lars@samba.org> (via SUSE),
823 Patch from Dirk Mueller <dmueller@suse.com>
825 PP/13 tls_peerdn now print-escaped for spool files.
826 Observed some $tls_peerdn in wild which contained \n, which resulted
827 in spool file corruption.
829 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
830 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
831 or write after TLS renegotiation, which otherwise led to messages
834 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
835 as a tracking header (ie: a signed header comes before the signature).
836 Patch from Wolfgang Breyha.
838 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
839 comma-sep list; embedded commas doubled.
841 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
843 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
845 Report and patch from Dmitry Banschikov.
847 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
848 Removed SSL_clear() after SSL_new() which led to protocol negotiation
849 failures. We appear to now support TLS1.1+ with Exim.
851 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
852 lets Exim select keys and certificates based upon TLS SNI from client.
853 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
854 before an outbound SMTP session. New log_selector, +tls_sni.
856 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
857 NULL dereference. Report and patch from Alun Jones.
859 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
860 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
861 Not seeing resolver debug output on NetBSD, but suspect this is a
862 resolver implementation change.
864 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
865 Left warnings. Added "eximon gdb" invocation mode.
867 PP/21 Defaulting "accept_8bitmime" to true, not false.
869 PP/22 Added -bw for inetd wait mode support.
871 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
872 locate the relevant includes and libraries. Made this the default.
874 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
875 Bugzilla 1246, report and most of solution from Tomasz Kusy.
877 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
878 This may cause build issues on older platforms.
880 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
881 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
882 gnutls_require_mac & gnutls_require_protocols (no longer supported).
883 Added SNI support via GnuTLS too.
884 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
886 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
888 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
889 Applied second patch from Janne, implementing suggestion to default
890 multiple-strings-in-record handling to match SPF spec.
892 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
894 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
895 read-only, out of scope).
896 Patch from Wolfgang Breyha, report from Stuart Northfield.
898 PP/29 Fix three issues highlighted by clang analyser static analysis.
899 Only crash-plausible issue would require the Cambridge-specific
900 iplookup router and a misconfiguration.
901 Report from Marcin Mirosław.
903 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
905 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
906 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
907 As part of this, removing so much warning spew let me fix some minor
908 real issues in debug logging.
910 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
911 assignment on my part. Fixed.
913 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
914 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
915 Janne Snabb (who went above and beyond: thank you).
917 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
918 string otherwise requires a connection and a bunch more work and it's
919 relatively easy to get wrong. Should also expose TLS library linkage
922 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
923 64-bit ${eval} (JH/03).
925 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
926 GNU libc to support some of the 64-bit stuff, should not lead to
927 conflicts. Defined before os.h is pulled in, so if a given platform
928 needs to override this, it can.
930 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
931 protection layer was required, which is not implemented.
932 Bugzilla 1254, patch from Wolfgang Breyha.
934 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
935 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
936 tls_dhparam take prime identifiers. Also unbreak combination of
937 OpenSSL+DH_params+TLSSNI.
939 PP/39 Disable SSLv2 by default in OpenSSL support.
945 PP/01 Solaris build fix for Oracle's LDAP libraries.
946 Bugzilla 1109, patch from Stephen Usher.
948 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
950 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
953 TF/02 Fix a couple more cases where we did not log the error message
954 when unlink() failed. See also change 4.74-TF/03.
956 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
957 lock up or crash if it happened to be inside a call to libc when it
958 got a SIGUSR1 from exiwhat.
960 The SIGUSR1 handler appends the current process status to the process
961 log which is later printed by exiwhat. It used to use the general
962 purpose logging code to do this, but several functions it calls are
963 not safe for signals.
965 The new output code in the SIGUSR1 handler is specific to the process
966 log, and simple enough that it's easy to inspect for signal safety.
967 Removing some special cases also simplifies the general logging code.
968 Removing the spurious timestamps from the process log simplifies
971 TF/04 Improved ratelimit ACL condition.
973 The /noupdate option has been deprecated in favour of /readonly which
974 has clearer semantics. The /leaky, /strict, and /readonly update modes
975 are mutually exclusive. The update mode is no longer included in the
976 database key; it just determines when the database is updated. (This
977 means that when you upgrde Exim will forget old rate measurements.)
979 Exim now checks that the per_* options are used with an update mode that
980 makes sense for the current ACL. For example, when Exim is processing a
981 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
982 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
983 must specify per_mail/readonly. If you omit the update mode it defaults to
984 /leaky where that makes sense (as before) or /readonly where required.
986 The /noupdate option is now undocumented but still supported for
987 backwards compatibility. It is equivalent to /readonly except that in
988 ACLs where /readonly is required you may specify /leaky/noupdate or
989 /strict/noupdate which are treated the same as /readonly.
991 A useful new feature is the /count= option. This is a generalization
992 of the per_byte option, so that you can measure the throughput of other
993 aggregate values. For example, the per_byte option is now equivalent
994 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
996 The per_rcpt option has been generalized using the /count= mechanism
997 (though it's more complicated than the per_byte equivalence). When it is
998 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
999 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1000 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1001 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1002 Note that using per_rcpt with a non-readonly update mode in more than
1003 one ACL will cause the recipients to be double-counted. (The per_mail
1004 and per_byte options don't have this problem.)
1006 The handling of very low rates has changed slightly. If the computed rate
1007 is less than the event's count (usually one) then this event is the first
1008 after a long gap. In this case the rate is set to the same as this event's
1009 count, so that the first message of a spam run is counted properly.
1011 The major new feature is a mechanism for counting the rate of unique
1012 events. The new per_addr option counts the number of different
1013 recipients that someone has sent messages to in the last time period. It
1014 behaves like per_rcpt if all the recipient addresses are different, but
1015 duplicate recipient addresses do not increase the measured rate. Like
1016 the /count= option this is a general mechanism, so the per_addr option
1017 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1018 example, measure the rate that a client uses different sender addresses
1019 with the options per_mail/unique=$sender_address. There are further
1020 details in the main documentation.
1022 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1024 TF/06 Removed a few PCRE remnants.
1026 TF/07 Automatically extract Exim's version number from tags in the git
1027 repository when doing development or release builds.
1029 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1030 Bugzilla 879. Patch from Paul Fisher.
1032 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1033 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1036 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1038 PP/05 Variable $av_failed, true if the AV scanner deferred.
1039 Bugzilla 1078. Patch from John Horne.
1041 PP/06 Stop make process more reliably on build failure.
1042 Bugzilla 1087. Patch from Heiko Schlittermann.
1044 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1045 Bugzilla 1089. Patch from Heiko Schlittermann.
1047 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1048 Bugzilla 1131. Patch from Holger Weiß.
1050 PP/09 Handle IPv6 addresses with SPF.
1051 Bugzilla 860. Patch from Wolfgang Breyha.
1053 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1055 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1058 PP/11 match_* no longer expand right-hand-side by default.
1059 New compile-time build option, EXPAND_LISTMATCH_RHS.
1060 New expansion conditions, "inlist", "inlisti".
1062 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1064 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1066 PP/14 fix log_write() format string regression from TF/03.
1067 Bugzilla 1152. Patch from Dmitry Isaikin.
1073 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1075 PP/02 Harmonised TLS library version reporting; only show if debugging.
1076 Layout now matches that introduced for other libraries in 4.74 PP/03.
1078 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1080 PP/04 New "dns_use_edns0" global option.
1082 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1085 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1086 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1088 TK/01 Updated PolarSSL code to 0.14.2.
1089 Bugzilla 1097. Patch from Andreas Metzler.
1091 PP/07 Catch divide-by-zero in ${eval:...}.
1092 Fixes bugzilla 1102.
1094 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1097 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1098 format-string attack -- SECURITY: remote arbitrary code execution.
1100 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1101 time unintentionally subject to list matching rules, letting the header
1102 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1103 arbitrary string expansion). This allowed for information disclosure.
1105 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1106 INT_MIN/-1 -- value coerced to INT_MAX.
1112 NM/01 Workround for PCRE version dependency in version reporting
1115 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1116 This fixes portability to compilers other than gcc, notably
1117 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1119 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1120 makefiles for portability to HP-UX and POSIX correctness.
1122 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1123 Also via indented variable definition in the Makefile.
1124 (Debugging by Oliver Heesakkers).
1126 PP/02 Restore caching of spamd results with expanded spamd_address.
1127 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1129 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1130 Improves build reliability. Fix from: Frank Elsner
1132 NM/02 Fix wide character breakage in the rfc2047 coding
1133 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1135 NM/03 Allow underscore in dnslist lookups
1136 Fixes bug 1026. Patch from Graeme Fowler
1138 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1139 Code patches from Adam Ciarcinski of NetBSD.
1141 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1144 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1145 is logged, to avoid truncation. Patch from John Horne.
1147 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1148 Patch from Jakob Hirsch.
1150 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1151 SQL string expansion failure details.
1152 Patch from Andrey Oktyabrski.
1154 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1155 Patch from Simon Arlott.
1157 PP/09 New lookups functionality failed to compile on old gcc which rejects
1158 extern declarations in function scope.
1159 Patch from Oliver Fleischmann
1161 PP/10 Use sig_atomic_t for flags set from signal handlers.
1162 Check getgroups() return and improve debugging.
1163 Fixed developed for diagnosis in bug 927 (which turned out to be
1166 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1167 Patch from Mark Zealey.
1169 PP/12 Bugzilla 1056: Improved spamd server selection.
1170 Patch from Mark Zealey.
1172 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1173 Based on patch from Heiko Schlittermann.
1175 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1176 Patch from Uwe Doering, sign-off by Michael Haardt.
1178 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1179 variable declaration deep within a block. Bug and patch from
1182 PP/15 lookups-Makefile IRIX compatibilty coercion.
1184 PP/16 Make DISABLE_DKIM build knob functional.
1186 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1187 Patch by Simon Arlott
1189 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1190 variable argument macros. Our copy now differs from upstream.
1196 TF/01 Failure to get a lock on a hints database can have serious
1197 consequences so log it to the panic log.
1199 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1200 controlled using the smtp_confirmation log selector.
1202 TF/03 Include the error message when we fail to unlink a spool file.
1204 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1205 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1206 for maintaining out-of-tree patches for some time.
1208 PP/01 Bugzilla 139: Documentation and portability issues.
1209 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1210 Handle per-OS dynamic-module compilation flags.
1212 PP/02 Let /dev/null have normal permissions.
1213 The 4.73 fixes were a little too stringent and complained about the
1214 permissions on /dev/null. Exempt it from some checks.
1215 Reported by Andreas M. Kirchwitz.
1217 PP/03 Report version information for many libraries, including
1218 Exim version information for dynamically loaded libraries. Created
1219 version.h, now support a version extension string for distributors
1220 who patch heavily. Dynamic module ABI change.
1222 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1223 privilege escalation vulnerability whereby the Exim run-time user
1224 can cause root to append content of the attacker's choosing to
1227 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1230 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1231 If dropping privileges for untrusted macros, we disabled normal logging
1232 on the basis that it would fail; for the Exim run-time user, this is not
1233 the case, and it resulted in successful deliveries going unlogged.
1234 Fixed. Reported by Andreas Metzler.
1240 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1241 only prepend for the Resent-* case. Fixes regression introduced in
1242 Exim 4.70 by NM/22 for Bugzilla 607.
1244 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1245 increasing numbers of administrators be bitten by this.
1247 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1249 PP/03 Bugzilla 994: added openssl_options main configuration option.
1251 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1253 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1255 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1257 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1258 without a peer certificate, leading to a segfault because of an
1259 assumption that peers always have certificates. Be a little more
1260 paranoid. Problem reported by Martin Tscholak.
1262 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1263 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1264 NB: ClamAV planning to remove STREAM in "middle of 2010".
1265 CL also introduces -bmalware, various -d+acl logging additions and
1266 more caution in buffer sizes.
1268 PP/09 Implemented reverse_ip expansion operator.
1270 PP/10 Bugzilla 937: provide a "debug" ACL control.
1272 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1274 PP/12 Bugzilla 973: Implement --version.
1276 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1278 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1280 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1282 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1283 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1284 ignore trailing whitespace.
1286 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1288 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1291 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1292 Notification from Dr Andrew Aitchison.
1294 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1295 ExtendedDetectionInfo response format.
1296 Notification from John Horne.
1298 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1301 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1302 XSL and documented dependency on system catalogs, with examples of how
1305 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1308 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1309 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1310 configuration file which is writeable by the Exim user or group.
1312 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1313 of configuration files to cover files specified with the -C option if
1314 they are going to be used with root privileges, not just the default
1317 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1318 option (effectively making it always true).
1320 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1321 files to be used while preserving root privileges.
1323 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1324 that rogue child processes cannot use them.
1326 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1327 run-time user, instead of root.
1329 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1330 Exim run-time user without dropping privileges.
1332 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1333 result string, instead of calling string_vformat() twice with the same
1336 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1337 for other users. Others should always drop root privileges if they use
1338 -C on the command line, even for a whitelisted configure file.
1340 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1342 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1348 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1349 $data_path, and $header_path variables; fixed documentation bugs and
1352 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1353 exipick to access non-standard spools, including the "frozen" queue
1356 NM/01 Bugzilla 965: Support mysql stored procedures.
1357 Patch from Alain Williams
1359 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1361 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1362 Patch from Andreas Metzler
1364 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1365 Patch from Kirill Miazine
1367 NM/05 Bugzilla 671: Added umask to procmail example.
1369 JJ/03 installed exipick 20100323.0, fixing doc bug
1371 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1372 directory. Notification and patch from Dan Rosenberg.
1374 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1376 TK/02 Improve log output when DKIM signing operation fails.
1378 MH/01 Treat the transport option dkim_domain as a colon separated
1379 list, not as a single string, and sign the message with each element,
1380 omitting multiple occurences of the same signer.
1382 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1383 Bugzilla 985, 986. Patch by Simon Arlott
1385 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1386 Patch by Simon Arlott
1388 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1389 MBX locking. Notification from Dan Rosenberg.
1395 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1397 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1399 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1401 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1403 NM/04 Bugzilla 915: Flag broken perl installation during build.
1409 TK/01 Added patch by Johannes Berg that expands the main option
1410 "spamd_address" if it starts with a dollar sign.
1412 TK/02 Write list of recipients to X-Envelope-Sender header when building
1413 the mbox-format spool file for content scanning (suggested by Jakob
1416 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1417 (http://www.dcc-servers.net/) support via dccifd. Activated by
1418 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1420 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1421 by Mark Daniel Reidel <mr@df.eu>.
1423 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1424 When building exim an external PCRE library is now needed -
1425 PCRE is a system library on the majority of modern systems.
1426 See entry on PCRE_LIBS in EDITME file.
1428 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1429 conversation. Added nologin parameter to request.
1430 Patch contributed by Kirill Miazine.
1432 TF/01 Do not log submission mode rewrites if they do not change the address.
1434 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1436 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1437 log files in place. Contributed by Roberto Lima.
1439 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1441 TF/03 Bugzilla 615: When checking the local_parts router precondition
1442 after a local_part_suffix or local_part_prefix option, Exim now
1443 does not use the address's named list lookup cache, since this
1444 contains cached lookups for the whole local part.
1446 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1447 Robert Millan. Documentation is in experimental-spec.txt.
1449 TF/04 Bugzilla 668: Fix parallel build (make -j).
1451 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1453 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1454 Patch provided by Jan Srzednicki.
1456 TF/05 Leading white space used to be stripped from $spam_report which
1457 wrecked the formatting. Now it is preserved.
1459 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1460 that they are available at delivery time.
1462 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1464 TF/08 TLS error reporting now respects the incoming_interface and
1465 incoming_port log selectors.
1467 TF/09 Produce a more useful error message if an SMTP transport's hosts
1468 setting expands to an empty string.
1470 NM/06 Bugzilla 744: EXPN did not work under TLS.
1471 Patch provided by Phil Pennock.
1473 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1474 Patch provided by Richard Godbee.
1476 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1477 acl_smtp_notquit, added index entry.
1479 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1480 Patch provided by Eugene Bujak.
1482 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1483 accept(). Patch provided by Maxim Dounin.
1485 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1486 Patch provided by Phil Pennock.
1488 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1490 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1491 Patch provided by Brad "anomie" Jorsch.
1493 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1494 Patch provided by Dean Brooks.
1496 TK/05 Add native DKIM support (does not depend on external libraries).
1498 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1499 Patch provided by Graeme Fowler.
1501 NM/16 Bugzilla 851: Documentation example syntax fix.
1503 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1505 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1508 NM/19 Bugzilla 745: TLS version reporting.
1509 Patch provided by Phil Pennock.
1511 NM/20 Bugzilla 167: bool: condition support.
1512 Patch provided by Phil Pennock.
1514 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1515 clients. Patch provided by Phil Pennock.
1517 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1518 Patch provided by Brad "anomie" Jorsch.
1520 NM/23 Bugzilla 687: Fix misparses in eximstats.
1521 Patch provided by Heiko Schlittermann.
1523 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1524 Patch provided by Heiko Schlittermann.
1526 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1527 plus update to original patch.
1529 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1531 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1532 Patch provided by David Brownlee.
1534 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1536 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1538 NM/30 Bugzilla 888: TLS documentation bugfixes.
1540 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1542 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1543 Unlike the original bugzilla I have changed all shell scripts in src tree.
1545 NM/33 Bugzilla 898: Transport filter timeout fix.
1546 Patch by Todd Rinaldo.
1548 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1549 Patch by Serge Demonchaux.
1551 NM/35 Bugzilla 39: Base64 decode bug fixes.
1552 Patch by Jakob Hirsch.
1554 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1556 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1558 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1560 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1566 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1567 ALT-N's libdkim that I have put here:
1568 http://duncanthrax.net/exim-experimental/
1570 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1571 were called 'true' and it seems that C99 defines that as a reserved
1572 keyword to be used with 'bool' variable types. That means you could
1573 not include C99-style headers which use bools without triggering
1574 build errors in sieve.c.
1576 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1577 as mailq or other aliases. Changed the --help handling significantly
1578 to do whats expected. exim_usage() emits usage/help information.
1580 SC/01 Added the -bylocaldomain option to eximstats.
1582 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1584 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1586 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1592 PH/01 Another patch from the Sieve maintainer.
1594 PH/02 When an IPv6 address is converted to a string for single-key lookup
1595 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1596 dots are used instead of colons so that keys in lsearch files need not
1597 contain colons. This was done some time before quoting was made available
1598 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1599 (notated using the quote facility) so as to distinguish them from IPv4
1600 keys. This meant that lookups for IP addresses in host lists did not work
1601 for iplsearch lookups.
1603 This has been fixed by arranging for IPv6 addresses to be expressed with
1604 colons if the lookup type is iplsearch. This is not incompatible, because
1605 previously such lookups could never work.
1607 The situation is now rather anomolous, since one *can* have colons in
1608 ordinary lsearch keys. However, making the change in all cases is
1609 incompatible and would probably break a number of configurations.
1611 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1614 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1615 conversion specification without a maximum field width, thereby enabling
1616 a rogue spamd server to cause a buffer overflow. While nobody in their
1617 right mind would setup Exim to query an untrusted spamd server, an
1618 attacker that gains access to a server running spamd could potentially
1619 exploit this vulnerability to run arbitrary code as the Exim user.
1621 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1622 $primary_hostname instead of what libspf2 thinks the hosts name is.
1624 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1625 a directory entry by the name of the lookup key. Previously, if a
1626 symlink pointed to a non-existing file or a file in a directory that
1627 Exim lacked permissions to read, a lookup for a key matching that
1628 symlink would fail. Now it is enough that a matching directory entry
1629 exists, symlink or not. (Bugzilla 503.)
1631 PH/03 The body_linecount and body_zerocount variables are now exported in the
1634 PH/04 Added the $dnslist_matched variable.
1636 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1637 This means they are set thereafter only if the connection becomes
1640 PH/06 Added the client_condition to authenticators so that some can be skipped
1641 by clients under certain conditions.
1643 PH/07 The error message for a badly-placed control=no_multiline_responses left
1644 "_responses" off the end of the name.
1646 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1648 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1649 (without spaces) instead of just copying the configuration text.
1651 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1653 PH/11 Added $max_received_linelength.
1655 PH/12 Added +ignore_defer and +include_defer to host lists.
1657 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1658 way in which PCRE > 7.0 is built.
1660 PH/14 Implemented queue_only_load_latch.
1662 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1663 MAIL command. The effect was to mangle the value on 64-bit systems.
1665 PH/16 Another patch from the Sieve maintainer.
1667 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1669 PH/18 If a system quota error occurred while trying to create the file for
1670 a maildir delivery, the message "Mailbox is full" was not appended to the
1671 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1672 only to a quota excession during the actual writing of the file.
1674 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1675 characters?) which causes problems in log lines. The DN values are now
1676 passed through string_printing() before being added to log lines.
1678 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1679 and InterBase are left for another time.)
1681 PH/21 Added message_body_newlines option.
1683 PH/22 Guard against possible overflow in moan_check_errorcopy().
1685 PH/23 POSIX allows open() to be a macro; guard against that.
1687 PH/24 If the recipient of an error message contained an @ in the local part
1688 (suitably quoted, of course), incorrect values were put in $domain and
1689 $local_part during the evaluation of errors_copy.
1695 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1696 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1699 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1700 issue a MAIL command.
1702 PH/02 In an ACL statement such as
1704 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1706 if a client was not listed at all, or was listed with a value other than
1707 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1708 the condition was not true (as it should be), so access was not denied.
1709 The bug was that the ! inversion was incorrectly passed on to the second
1710 item. This has been fixed.
1712 PH/03 Added additional dnslists conditions == and =& which are different from
1713 = and & when the dns lookup returns more than one IP address.
1715 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1716 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1718 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1719 FSYNC, which compiles an option called disable_fsync that allows for
1720 bypassing fsync(). The documentation is heavily laced with warnings.
1722 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1724 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1725 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1726 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1727 including adding "make clean"; (3) Added -fPIC when compiling the test
1728 dynamically loaded module, to get rid of a warning.
1730 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1731 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1732 = 0s. The bug is otherwise harmless.
1734 PH/07 There was a bug in the dovecot authenticator such that the value of
1735 $auth1 could be overwritten, and so not correctly preserved, after a
1736 successful authentication. This usually meant that the value preserved by
1737 the server_setid option was incorrect.
1739 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1741 PH/09 Installed PCRE release 7.0.
1743 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1744 run for batched SMTP input. It is now run at the start of every message
1745 in the batch. While fixing this I discovered that the process information
1746 (output by running exiwhat) was not always getting set for -bs and -bS
1747 input. This is fixed, and it now also says "batched" for BSMTP.
1749 PH/11 Added control=no_pipelining.
1751 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1752 patch, slightly modified), and move the expansion of helo_data till after
1753 the connection is made in the smtp transport (so it can use these
1756 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1758 PH/14 Added log_selector = +pid.
1760 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1762 PH/16 Add ${if forany and ${if forall.
1764 PH/17 Added dsn_from option to vary the From: line in DSNs.
1766 PH/18 Flush SMTP output before performing a callout, unless control =
1767 no_callout_flush is set.
1769 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1770 was true (the default) a successful delivery failed to delete the retry
1771 item, thus causing premature timeout of the address. The bug is now
1774 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1776 PH/21 Long custom messages for fakedefer and fakereject are now split up
1777 into multiline reponses in the same way that messages for "deny" and
1778 other ACL rejections are.
1780 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1781 with slight modification.
1783 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1784 draft, changing the syntax and factoring some duplicate code".
1786 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1787 for deliveries of the second and subsequent messages over the same SMTP
1790 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1791 ${reduce, with only minor "tidies".
1793 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1795 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1796 expansion side effects.
1798 PH/27 When a message times out after an over-quota error from an Exim-imposed
1799 quota, the bounce message says "mailbox is full". This message was not
1800 being given when it was a system quota that was exceeded. It now should
1803 MH/03 Made $recipients available in local_scan(). local_scan() already has
1804 better access to the recipient list through recipients_list[], but
1805 $recipients can be useful in postmaster-provided expansion strings.
1807 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1808 in the case of a MAIL command with additional options following the
1809 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1810 were accidentally chopped off.
1812 PH/29 SMTP synchronization checks are implemented when a command is read -
1813 there is a check that no more input is waiting when there shouldn't be
1814 any. However, for some commands, a delay in an ACL can mean that it is
1815 some time before the response is written. In this time, more input might
1816 arrive, invalidly. So now there are extra checks after an ACL has run for
1817 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1818 pipelining has not been advertised.
1820 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1822 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1823 returned in the SMTP response when smtp_return_error_details was false.
1824 This has been fixed.
1826 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1827 instead of the C I/O that was originally supplied, because problems were
1828 reported on Solaris.
1830 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1831 Exim which did not show up earlier: it was assuming that a call to
1832 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1833 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1834 was a macro that became an assignment, so it seemed to work. This has
1835 changed to a proper function call with a void return, hence the compile
1836 error. Exim's code has been fixed.
1838 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1841 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1843 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1845 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1846 because it turns out to be too big or there is a timeout) there is no
1847 "Completed" line in the log. When some messages of this type were
1848 selected by exigrep, they were listed as "not completed". Others were
1849 picked up by some special patterns. I have improved the selection
1850 criteria to be more general.
1852 PH/38 The host_find_failed option in the manualroute router can now be set
1853 to "ignore", to completely ignore a host whose IP address cannot be
1854 found. If all hosts are ignored, the behaviour is controlled by the new
1855 host_all_ignored option.
1857 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1858 homing or because of multiple MX records with /mx) generated more than
1859 one IP address, and the following item turned out to be the local host,
1860 all the secondary addresses of the first item were incorrectly removed
1861 from the list, along with the local host and any following hosts (which
1862 is what is supposed to happen).
1864 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1865 whoever called Exim into the -H file. In the case of the daemon it was
1866 behaving confusingly. When first started, it used values for whoever
1867 started the daemon, but after a SIGHUP it used the Exim user (because it
1868 calls itself on a restart). I have changed the code so that it now always
1871 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1872 message are rejected with the same error (e.g. no authentication or bad
1873 sender address), and a DATA command is nevertheless sent (as can happen
1874 with PIPELINING or a stupid MUA), the error message that was given to the
1875 RCPT commands is included in the rejection of the DATA command. This is
1876 intended to be helpful for MUAs that show only the final error to their
1879 PH/42 Another patch from the Sieve maintainer.
1881 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1882 Eximstats - Fixed some broken HTML links and added missing column headers
1884 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1885 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1887 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1889 PH/43 Yet another patch from the Sieve maintainer.
1891 PH/44 I found a way to check for a TCP/IP connection going away before sending
1892 the response to the final '.' that terminates a message, but only in the
1893 case where the client has not sent further data following the '.'
1894 (unfortunately, this is allowed). However, in many cases there won't be
1895 any further data because there won't be any more messages to send. A call
1896 to select() can be used: if it shows that the input is "ready", there is
1897 either input waiting, or the socket has been closed. An attempt to read
1898 the next input character can distinguish the two cases. Previously, Exim
1899 would have sent an OK response which the client would never have see.
1900 This could lead to message repetition. This fix should cure that, at
1901 least in a lot of common cases.
1903 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1904 advertised in response to EHLO.
1910 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1911 fixed by 4.65/MH/01 (is this a record?) are fixed:
1913 (i) An empty string was always treated as zero by the numeric comparison
1914 operators. This behaviour has been restored.
1916 (ii) It is documented that the numeric comparison operators always treat
1917 their arguments as decimal numbers. This was broken in that numbers
1918 starting with 0 were being interpreted as octal.
1920 While fixing these problems I realized that there was another issue that
1921 hadn't been noticed. Values of message_size_limit (both the global option
1922 and the transport option) were treated as octal if they started with 0.
1923 The documentation was vague. These values are now always treated as
1924 decimal, and I will make that clear in the documentation.
1930 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1931 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1934 MH/01 Don't check that the operands of numeric comparison operators are
1935 integers when their expansion is in "skipping" mode (fixes bug
1936 introduced by 4.64-PH/07).
1938 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1939 child addresses, Exim now panics and dies. Previously, because the count
1940 is held in a short int, deliveries were likely to be lost. As such a
1941 large number of recipients for a single message is ridiculous
1942 (performance will be very, very poor), I have chosen to impose a limit
1943 rather than extend the field.
1949 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1950 leftover -K file (the existence of which was triggered by #402).
1951 While we were at it, introduced process PID as part of the -K
1952 filename. This should rule out race conditions when creating
1955 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1956 processing considerably. Previous code took too long for large mails,
1957 triggering a timeout which in turn triggers #401.
1959 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1960 in the DK code in transports.c. sendfile() is not really portable,
1961 hence the _LINUX specificness.
1963 TF/01 In the add_headers option to the mail command in an Exim filter,
1964 there was a bug that Exim would claim a syntax error in any
1965 header after the first one which had an odd number of characters
1968 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1969 callout verification, Exim cached a "reject" for the entire domain. This
1970 is correct for most verifications, but it is not correct for a recipient
1971 verification with use_sender or use_postmaster set, because in that case
1972 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1973 case of MAIL FROM:<> rejection from other early rejections (e.g.
1974 rejection of HELO). When verifying a recipient using a non-null MAIL
1975 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1976 Whatever the result of the callout, the value of the domain cache is
1977 left unchanged (for any other kind of callout, getting as far as trying
1978 RCPT means that the domain itself is ok).
1980 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1983 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1984 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1985 session, the daemon ignores ECONNECT errors and logs others; it now
1986 ignores EPIPE as well.
1988 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1989 (quoted-printable decoding).
1991 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1992 later the small subsequent patch to fix an introduced bug.
1994 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1996 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1998 PH/08 An error is now given if message_size_limit is specified negative.
2000 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2001 to be given (somewhat) arbitrary names.
2003 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2006 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2007 miscellaneous code fixes
2009 PH/10 Added the log_reject_target ACL modifier to specify where to log
2012 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2013 hostname. This is wrong, because it relates to the incoming message (and
2014 probably the interface on which it is arriving) and not to the outgoing
2015 callout (which could be using a different interface). This has been
2016 changed to use the value of the helo_data option from the smtp transport
2017 instead - this is what is used when a message is actually being sent. If
2018 there is no remote transport (possible with a router that sets up host
2019 addresses), $smtp_active_hostname is used.
2021 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2022 tweaks were necessary in order to get it to work (see also 21 below):
2023 (a) The code assumed that strncpy() returns a negative number on buffer
2024 overflow, which isn't the case. Replaced with Exim's string_format()
2026 (b) There were several signed/unsigned issues. I just did the minimum
2027 hacking in of casts. There is scope for a larger refactoring.
2028 (c) The code used strcasecmp() which is not a standard C function.
2029 Replaced with Exim's strcmpic() function.
2030 (d) The code set only $1; it now sets $auth1 as well.
2031 (e) A simple test gave the error "authentication client didn't specify
2032 service in request". It would seem that Dovecot has changed its
2033 interface. Fortunately there's a specification; I followed it and
2034 changed what the client sends and it appears to be working now.
2036 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2039 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2040 address A is aliased to B and C, where B exists and C does not. Without
2041 -v the output is "A verified" because verification stops after a
2042 successful redirection if more than one address is generated. However,
2043 with -v the child addresses are also verified. Exim was outputting "A
2044 failed to verify" and then showing the successful verification for C,
2045 with its parentage. It now outputs "B failed to verify", showing B's
2046 parentage before showing the successful verification of C.
2048 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2049 look up a TXT record in a specific list after matching in a combined
2052 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2053 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2054 they consult the DNS. I had assumed they would set it the way they
2055 wanted; and indeed my experiments on Linux seem to show that in some
2056 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2057 To be on the safe side, however, I have now made the interface to
2058 host_find_byname() similar to host_find_bydns(), with an argument
2059 containing the DNS resolver options. The host_find_byname() function now
2060 sets these options at its start, just as host_find_bydns() does. The smtp
2061 transport options dns_qualify_single and dns_search_parents are passed to
2062 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2063 of host_find_byname() use the default settings of RES_DEFNAMES
2064 (qualify_single) but not RES_DNSRCH (search_parents).
2066 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2067 spool_read_header() do less string testing, by means of a preliminary
2068 switch on the second character of optional "-foo" lines. (This is
2069 overdue, caused by the large number of possibilities that now exist.
2070 Originally there were few.) While I was there, I also converted the
2071 str(n)cmp tests so they don't re-test the leading "-" and the first
2072 character, in the hope this might squeeze out yet more improvement.
2074 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2075 flag allowing group syntax was set by the header_syntax check but not
2076 turned off, possible causing trouble later; (2) The flag was not being
2077 set at all for the header_verify test, causing "group"-style headers to
2078 be rejected. I have now set it in this case, and also caused header_
2079 verify to ignore an empty address taken from a group. While doing this, I
2080 came across some other cases where the code for allowing group syntax
2081 while scanning a header line wasn't quite right (mostly, not resetting
2082 the flag correctly in the right place). These bugs could have caused
2083 trouble for malformed header lines. I hope it is now all correct.
2085 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2086 with the "reply" argument non-NULL. The code, however (which originally
2087 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2088 but it didn't always do it. This confused somebody who was copying the
2089 code for some other use. I have removed all the tests.
2091 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2092 feature that was used to support insecure browsers during the U.S. crypto
2093 embargo. It requires special client support, and Exim is probably the
2094 only MTA that supported it -- and would never use it because real RSA is
2095 always available. This code has been removed, because it had the bad
2096 effect of slowing Exim down by computing (never used) parameters for the
2097 RSA_EXPORT functionality.
2099 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2100 authenticator to fail if there's a tab character in the incoming data
2101 (there should never be unless someone is messing about, as it's supposed
2102 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2103 if the connection is using TLS or if the remote IP is the same as the
2104 local IP, and the "valid-client-cert option" if a client certificate has
2107 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2108 authenticators. This can be used for authorization after authentication
2109 succeeds. (In the case of plaintext, it servers for both authentication
2112 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2113 if any retry times were supplied.
2115 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2116 connection, where there is no client IP address to check. In this
2117 situation, the verify now always succeeds.
2119 PH/25 Applied John Jetmore's -Mset patch.
2121 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2123 PH/27 In a string expansion for a processed (not raw) header when multiple
2124 headers of the same name were present, leading whitespace was being
2125 removed from all of them, but trailing whitespace was being removed only
2126 from the last one. Now trailing whitespace is removed from each header
2127 before concatenation. Completely empty headers in a concatenation (as
2128 before) are ignored.
2130 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2131 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2133 PH/29 [Removed. This was a change that I later backed out, and forgot to
2134 correct the ChangeLog entry (that I had efficiently created) before
2135 committing the later change.]
2137 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2138 address errors (4xx response to RCPT) over the same connection as other
2139 messages routed to the same hosts. Such deliveries are always "forced",
2140 so retry times are not inspected. This resulted in far too many retries
2141 for the affected addresses. The effect occurred only when there were more
2142 hosts than the hosts_max_try setting in the smtp transport when it had
2143 the 4xx errors. Those hosts that it had tried were not added to the list
2144 of hosts for which the message was waiting, so if all were tried, there
2145 was no problem. Two fixes have been applied:
2147 (i) If there are any address or message errors in an SMTP delivery, none
2148 of the hosts (tried or untried) are now added to the list of hosts
2149 for which the message is waiting, so the message should not be a
2150 candidate for sending over the same connection that was used for a
2151 successful delivery of some other message. This seems entirely
2152 reasonable: after all the message is NOT "waiting for some host".
2153 This is so "obvious" that I'm not sure why it wasn't done
2154 previously. Hope I haven't missed anything, but it can't do any
2155 harm, as the worst effect is to miss an optimization.
2157 (ii) If, despite (i), such a delivery is accidentally attempted, the
2158 routing retry time is respected, so at least it doesn't keep
2159 hammering the server.
2161 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2162 in ${readsocket because some servers need this prod.
2164 PH/32 Added some extra debug output when updating a wait-xxx database.
2166 PH/33 The hint "could be header name not terminated by colon", which has been
2167 given for certain expansion errors for a long time, was not being given
2168 for the ${if def:h_colon_omitted{... case.
2170 PH/34 The spec says: "With one important exception, whenever a domain list is
2171 being scanned, $domain contains the subject domain." There was at least
2172 one case where this was not true.
2174 PH/35 The error "getsockname() failed: connection reset by peer" was being
2175 written to the panic log as well as the main log, but it isn't really
2176 panic-worthy as it just means the connection died rather early on. I have
2177 removed the panic log writing for the ECONNRESET error when getsockname()
2180 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2181 runs only) independently of the message's sender address. This meant
2182 that, if the 4xx error was in fact related to the sender, a different
2183 message to the same recipient with a different sender could confuse
2184 things. In particualar, this can happen when sending to a greylisting
2185 server, but other circumstances could also provoke similar problems.
2186 I have changed the default so that the retry time for these errors is now
2187 based a combination of the sender and recipient addresses. This change
2188 can be overridden by setting address_retry_include_sender=false in the
2191 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2192 remote server are returned as part of bounce messages. This was not
2193 happening for LMTP over a pipe (the lmtp transport), but now it is the
2194 same for both kinds of LMTP.
2196 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2197 in header lines that were in fact CNAMEs. This is no longer the case.
2199 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2200 and queue runs started by the daemon processed all messages. This has
2201 been fixed so that -R and -S can now usefully be given with -q<time>.
2203 PH/40 Import PCRE release 6.7 (fixes some bugs).
2205 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2207 PH/42 Give an error if -q is specified more than once.
2209 PH/43 Renamed the variables $interface_address and $interface_port as
2210 $received_ip_address and $received_port, to make it clear that these
2211 values apply to message reception, and not to the outgoing interface when
2212 a message is delivered. (The old names remain recognized, of course.)
2214 PH/44 There was no timeout on the connect() call when using a Unix domain
2215 socket in the ${readsocket expansion. There now is.
2217 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2218 be meaningful with "accept".
2220 SC/01 Eximstats V1.43
2221 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2223 SC/02 Eximstats V1.44
2224 Use a glob alias rather than an array ref in the generated
2225 parser. This improves both readability and performance.
2227 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2228 Collect SpamAssassin and rejection statistics.
2229 Don't display local sender or destination tables unless
2230 there is data to show.
2231 Added average volumes into the top table text output.
2233 SC/04 Eximstats V1.46
2234 Collect data on the number of addresses (recipients)
2235 as well as the number of messages.
2237 SC/05 Eximstats V1.47
2238 Added 'Message too big' to the list of mail rejection
2239 reasons (thanks to Marco Gaiarin).
2241 SC/06 Eximstats V1.48
2242 Mainlog lines which have GMT offsets and are too short to
2243 have a flag are now skipped.
2245 SC/07 Eximstats V1.49 (Alain Williams)
2246 Added the -emptyok flag.
2248 SC/08 Eximstats V1.50
2249 Fixes for obtaining the IP address from reject messages.
2251 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2252 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2253 whitesspace changes from 4.64-PH/27
2255 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2258 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2259 are found, allow negative numbers in numeric criteria)
2261 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2263 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2264 to match changes made in 4.64-PH/43
2266 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2268 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2269 "domain" is actually an IP address, and force a failure. This locks out
2270 those revolvers/nameservers that support "A-for-A" lookups, in
2271 contravention of the specifications.
2273 PH/48 When a host name was looked up from an IP address, and the subsequent
2274 forward lookup of the name timed out, the host name was left in
2275 $sender_host_name, contrary to the specification.
2277 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2278 restricted to single-key lookups, Exim was not diagnosing an error if
2279 * or *@ was used with a query-style lookup.
2281 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2283 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2284 long ago, but noone interested enough thought of it. Let's just say that
2285 the "1.1" means that there are some new functions that weren't there at
2286 some point in the past.
2288 PH/51 Error processing for expansion failure of helo_data from an smtp
2289 transport during callout processing was broken.
2291 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2292 tested/used via the -bh/-bhc/-bs options.
2294 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2295 bug, fixed in subsequent PCRE releases).
2297 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2298 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2300 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2306 SC/01 Use a glob alias rather than an array ref in eximstats generated
2307 parser. This improves both readability and performance.
2309 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2310 Don't display local sender or destination tables in eximstats unless
2311 there is data to show.
2312 Added average volumes into the eximstats top table text output.
2314 SC/03 Collect data on the number of addresses (recipients) as well
2315 as the number of messages in eximstats.
2317 TF/01 Correct an error in the documentation for the redirect router. Exim
2318 does (usually) call initgroups() when daemonizing.
2320 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2321 with consistent privilege compared to when running as a daemon.
2323 TF/03 Note in the spec that $authenticated_id is not set for local
2324 submissions from trusted users.
2326 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2327 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2329 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2330 by adding some example configuration directives to the default
2331 configuration file. A little bit of work is required to uncomment the
2332 directives and define how usernames and passwords are checked, but
2333 there is now a framework to start from.
2335 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2336 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2337 without this. I don't know how relevant this is to other LDAP libraries.
2339 PH/02 Add the verb name to the "unknown ACL verb" error.
2341 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2343 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2345 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2346 directory not expanded when it should be if an expanded home directory
2347 was set for the address (which is overridden by the transport).
2349 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2352 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2353 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2354 because it is too late at that time, and has no effect.
2356 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2357 security issue with \' (bugzilla #107). I could not use the
2358 PQescapeStringConn() function, because it needs a PGconn value as one of
2361 PH/08 When testing addresses using -bt, indicate those final addresses that
2362 are duplicates that would not cause an additional delivery. At least one
2363 person was confused, thinking that -bt output corresponded to deliveries.
2364 (Suppressing duplicates isn't a good idea as you lose the information
2365 about possibly different redirections that led to the duplicates.)
2367 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2368 systems where poll() doesn't work, in particular OS X.
2370 PH/10 Added more information to debugging output for retry time not reached.
2372 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2373 operations in malware.c.
2375 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2378 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2379 syslog_duplication set false, log lines that would normally be written
2380 both the the main log and to the reject log were not written to syslog at
2383 PH/14 In the default configuration, change the use of "message" in ACL warn
2384 statements to "add_header".
2386 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2387 not followed by a command (e.g. "seen endif").
2389 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2390 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2393 PH/17 Added extra conditions to the default value of delay_warning_condition
2397 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2398 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2399 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2402 The Auto-Submitted: and various List- headers are standardised, whereas I
2403 don't think Precedence: ever was.
2405 PH/18 Refactored debugging code in route_finduser() to show more information,
2406 in particular, the error code if getpwnam() issues one.
2408 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2409 This is apparently needed in addition to the PH/07 change above to avoid
2410 any possible encoding problems.
2412 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2413 but not after initializing Perl.
2415 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2416 output them only if debugging. By default they are written stderr,
2417 apparently, which is not desirable.
2419 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2422 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2425 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2427 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2428 authenticated or an ident call has been made. Suppress the default
2429 values for $authenticated_id and $authenticated_sender (but permit -oMai
2430 and -oMas) when testing with -bh.
2432 PH/24 Re-jigged the order of the tests in the default configuration so that the
2433 tests for valid domains and recipients precede the DNS black list and CSA
2434 tests, on the grounds that those ones are more expensive.
2436 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2437 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2438 This bug exists in every version of Exim that I still have, right back to
2441 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2442 However, an attempt to turn on case-sensitivity in a regex key by
2443 including (?-i) didn't work because the subject string was already
2444 lowercased, and the effects were non-intuitive. It turns out that a
2445 one-line patch can be used to allow (?-i) to work as expected.
2451 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2452 other effects) broke the use of negated acl sub-conditions.
2454 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2457 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2458 "Deny" causes Exim to reject the incoming connection with a 554 error.
2459 Unfortunately, if there is a major crisis, such as a disk failure,
2460 tcp-wrappers gives "deny", whereas what one would like would be some
2461 kind of temporary error. A kludge has been added to help with this.
2462 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2463 554 error is used if errno is still zero or contains ENOENT (which occurs
2464 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2467 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2469 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2470 errors. Otherwise a message that provokes a temporary error (when other
2471 messages do not) can cause a whole host to time out.
2473 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2474 addresses were routed directly to files or pipes from a redirect router.
2475 File deliveries just didn't batch; pipe deliveries might have suffered
2478 PH/06 A failure to get a lock for a hints database would erroneously always say
2479 "Failed to get write lock", even when it was really a read lock.
2481 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2482 of 0600. This has been changed to use the value of the lockfile_mode
2483 option (which defaults to 0600).
2485 PH/08 Applied small patch from the Sieve maintainer.
2487 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2488 folder from quota calculations, a direct delivery into this folder messed
2489 up the contents of the maildirsize file. This was because the regex was
2490 used only to exclude .Trash (or whatever) when the size of the mailbox
2491 was calculated. There was no check that a delivery was happening into an
2492 excluded directory. This bug has been fixed by ignoring all quota
2493 processing for deliveries into excluded directories.
2495 PH/10 Added the maildirfolder_create_regex option to appendfile.
2501 PH/01 The code for finding all the local interface addresses on a FreeBSD
2502 system running IPv6 was broken. This may well have applied to all BSD
2503 systems, as well as to others that have similar system calls. The broken
2504 code found IPv4 interfaces correctly, but gave incorrect values for the
2505 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2506 that it would not match correctly against @[] and not recognize the IPv6
2509 PH/02 The ipliteral router was not recognizing addresses of the form user@
2510 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2512 PH/03 Added disable_ipv6.
2514 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2515 decoded form, because it is most often used to construct To: headers
2516 lines in autoreplies, and the decoded form may well be syntactically
2517 invalid. However, $reply_address has leading white space removed, and all
2518 newlines turned into spaces so that the autoreply transport does not
2521 PH/05 If group was specified without a user on a router, and no group or user
2522 was specified on a transport, the group from the router was ignored.
2524 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2525 for visible compile-time settings that can be used to change these
2526 numbers, for those that want even more. Backwards compatibility with old
2527 spool files has been maintained. However, going back to a previous Exim
2528 release will lost any variables that are in spool files.
2530 PH/07 Two small changes when running in the test harness: increase delay when
2531 passing a TCP/IP connection to a new process, in case the original
2532 process has to generate a bounce, and remove special handling of
2533 127.0.0.2 (sic), which is no longer necessary.
2535 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2536 be the same on different OS.
2538 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2541 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2542 whether --show-vars was specified or not
2544 JJ/02 exipick: Added support for new ACL variable spool format introduced
2547 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2548 syntactically invalid From: or Reply-to: line, and a filter used this to
2549 generate an autoreply, and therefore failed to obtain an address for the
2550 autoreply, Exim could try to deliver to a non-existent relative file
2551 name, causing unrelated and misleading errors. What now happens is that
2552 it logs this as a hard delivery error, but does not attempt to create a
2555 PH/11 The exinext utility has a -C option for testing purposes, but although
2556 the given file was scanned by exinext itself; it wasn't being passed on
2557 when Exim was called.
2559 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2560 an end-of-file indication when reading a command response.
2562 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2563 compiled. In many other places in Exim, IPv6 addresses are always
2564 recognized, so I have changed this. It also means that IPv4 domain
2565 literals of the form [IPV4:n.n.n.n] are now always recognized.
2567 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2568 used if the router is not running as root, for example, when verifying at
2569 ACL time, or when using -bh. The debugging output from this situation was
2570 non-existent - all you got was a failure to exec. I have made two
2573 (a) Failures to set uid/gid, the current directory, or a process leader
2574 in a subprocess such as that created by queryprogram now generate
2575 suitable debugging ouput when -d is set.
2577 (b) The queryprogram router detects when it is not running as root,
2578 outputs suitable debugging information if -d is set, and then runs
2579 the subprocess without attempting to change uid/gid.
2581 PH/15 Minor change to Makefile for building test_host (undocumented testing
2584 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2585 additional section of a DNS packet that returns MX or SRV records.
2586 Instead, it always explicitly searches for A/AAAA records. This avoids
2587 major problems that occur when a DNS server includes only records of one
2588 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2589 fixed another bug: if SRV records were looked up and the corresponding
2590 address records were *not* found in the additional section, the port
2591 values from the SRV records were lost.
2593 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2594 using the correct key (the original address) when searching the retry
2595 rules in order to find which one to use for generating the retry hint.
2597 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2598 adding the default one. Similarly, if it contains a Reply-To: header, the
2599 errors_reply_to option, if set, is not used.
2601 PH/19 When calculating a retry time, Exim used to measure the "time since
2602 failure" by looking at the "first failed" field in the retry record. Now
2603 it does not use this if it is later than than the arrival time of the
2604 message. Instead it uses the arrival time. This makes for better
2605 behaviour in cases where some deliveries succeed, thus re-setting the
2606 "first failed" field. An example is a quota failure for a huge message
2607 when small messages continue to be delivered. Without this change, the
2608 "time since failure" will always be short, possible causing more frequent
2609 delivery attempts for the huge message than are intended.
2610 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2612 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2613 $1, $2, $3) because the numerical variables can be reset during some
2614 expansion items (e.g. "match"), thereby losing the authentication data.
2616 PH/21 Make -bV show the size of off_t variables so that the test suite can
2617 decide whether to run tests for quotas > 2G.
2619 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2620 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2621 filecount value is greater than 2G or if a quota value is greater than 2G
2622 on a system where the size of off_t is not greater than 4, a panic error
2625 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2626 never match. The debug and -bh output now contains an explicit error
2627 message indicating a malformed IPv4 address or mask.
2629 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2630 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2631 PH/23 above applies.
2633 PH/25 Do not write to syslog when running in the test harness. The only
2634 occasion when this arises is a failure to open the main or panic logs
2635 (for which there is an explicit test).
2637 PH/26 Added the /no_tell option to "control=freeze".
2639 PH/27 If a host name lookup failed very early in a connection, for example, if
2640 the IP address matched host_lookup and the reverse lookup yielded a name
2641 that did not have a forward lookup, an error message of the form "no IP
2642 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2643 could be logged. Now it outputs the IP address instead of "NULL".
2645 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2646 allows the sender and the authenticated sender to be set when
2647 submitting a message from within Exim. Since child_open_exim() is
2648 documented for local_scan(), the new function should be too.
2650 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2651 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2652 results in an empty string is now treated as unset.
2654 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2656 PH/31 Added sender_verify_fail logging option.
2658 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2659 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2660 tidied the source and removed it altogether.
2662 PH/33 When a queue run was abandoned because the load average was too high, a
2663 log line was always written; now it is written only if the queue_run log
2664 selector is set. In addition, the log line for abandonment now contains
2665 information about the queue run such as the pid. This is always present
2666 in "start" and "stop" lines but was omitted from the "abandon" line.
2668 PH/34 Omit spaces between a header name and the colon in the error message that
2669 is given when verify = headers_syntax fails (if there are lots of them,
2670 the message gets confusing).
2672 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2673 names, as there are now some PTR records that contain slashes. This check
2674 is only to protect against broken name servers that fall over on strange
2675 characters, so the fact that it applies to all lookups doesn't matter.
2677 PH/36 Now that the new test suite is complete, we can remove some of the
2678 special code in Exim that was needed for the old test suite. For example,
2679 sorting DNS records because real resolvers return them in an arbitrary
2680 order. The new test suite's fake resolver always returns records in the
2683 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2684 bounces) except when queue_only is set, to avoid logging races between
2685 the different processes.
2687 PH/38 Panic-die if .include specifies a non-absolute path.
2689 PH/39 A tweak to the "H" retry rule from its user.
2691 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2692 a label. They prevented compilation on older perls.
2694 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2695 a warning to be raised on newish perls.
2697 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2698 on queue. Changes to match documented behaviour of showing count of
2699 messages matching specified criteria.
2701 PH/40 Changed the default ident timeout from 30s to 5s.
2703 PH/41 Added support for the use of login_cap features, on those BSD systems
2704 that have them, for controlling the resources used by pipe deliveries.
2706 PH/42 The content-scanning code uses fopen() to create files in which to put
2707 message data. Previously it was not paying any attention to the mode of
2708 the files. Exim runs with umask(0) because the rest of the code creates
2709 files with open(), and sets the required mode explicitly. Thus, these
2710 files were ending up world-writeable. This was not a big issue, because,
2711 being within the spool directory, they were not world-accessible. I have
2712 created a function called modefopen, which takes an additional mode
2713 argument. It sets umask(777), creates the file, chmods it to the required
2714 mode, then resets the umask. All the relevant calls to fopen() in the
2715 content scanning code have been changed to use this function.
2717 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2718 to 24 hours. This avoids potential overflow problems when processing G
2719 and H retry rules. I suspect nobody ever tinkers with this value.
2721 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2723 PH/45 When the plaintext authenticator is running as a client, the server's
2724 challenges are checked to ensure they are valid base64 strings. By
2725 default, the authentication attempt is cancelled if an invalid string is
2726 received. Setting client_ignore_invalid_base64 true ignores these errors.
2727 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2728 they are received. Thus, the responses can be made to depend on the
2729 challenges. If an invalid string is ignored, an empty string is placed in
2732 PH/46 Messages that are created by the autoreply transport now contains a
2733 References: header, in accordance with RFCs 2822 and 3834.
2735 PH/47 Added authenticated_sender_force to the smtp transport.
2737 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2739 PH/49 Installed latest patch from the Sieve maintainer.
2741 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2742 was also set, the appendfile transport was unnecessarily scanning a
2743 directory of message files (e.g. for maildir delivery) to find the count
2744 of files (along with the size), even though it did not need this
2745 information. It now does the scan only if it needs to find either the
2746 size of the count of files.
2748 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2750 PH/52 Two bugs concerned with error handling when the smtp transport is
2753 (i) Exim was not creating retry information for temporary errors given
2754 for individual recipients after the DATA command when the smtp transport
2755 was used in LMTP mode. This meant that they could be retried too
2756 frequently, and not timed out correctly.
2758 (ii) Exim was setting the flag that allows error details to be returned
2759 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2760 recipients that were returned after the DATA command.
2762 PH/53 This is related to PH/52, but is more general: for any failing address,
2763 when detailed error information was permitted to be returned to the
2764 sender, but the error was temporary, then after the final timeout, only
2765 "retry timeout exceeded" was returned. Now it returns the full error as
2766 well as "retry timeout exceeded".
2768 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2769 do this, and (what is worse) MTAs that accept it.
2771 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2772 will now be deprecated.
2774 PH/56 New os.c-cygwin from the Cygwin maintainer.
2776 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2777 formats (previously only available in exim formats via -bpr, -bpru,
2778 and -bpra. Now also available in native and exiqgrep formats)
2780 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2781 with very large, slow to parse queues
2783 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2785 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2787 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2788 responses to authentication challenges, though it was showing the
2789 challenges; (ii) I've removed the CR characters from the debug output for
2792 PH/58 Allow for the insertion of a newline as well as a space when a string
2793 is turned into more than one encoded-word during RFC 2047 encoding. The
2794 Sieve code now uses this.
2796 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2797 data_4xx, lost_connection, tls_required.
2799 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2800 message was being sent as an SMTP response.
2802 PH/61 Add -l and -k options to exicyclog.
2804 PH/62 When verifying, if an address was redirected to one new address, so that
2805 verification continued, and the new address failed or deferred after
2806 having set something in $address_data, the value of $address_data was not
2807 passed back to the ACL. This was different to the case when no
2808 redirection occurred. The value is now passed back in both cases.
2810 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2811 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2812 use login_cap.h, so on its own it isn't the distinguishing feature. The
2813 new name refers directly to the setclassresources() function.
2815 PH/65 Added configuration files for NetBSD3.
2817 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2819 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2820 is preferred over IPv4.
2822 PH/68 The bounce_return_message and bounce_return_body options were not being
2823 honoured for bounces generated during the reception of non-SMTP messages.
2824 In particular, this applied to messages rejected by the ACL. This bug has
2825 been fixed. However, if bounce_return_message is true and bounce_return_
2826 body is false, the headers that are returned for a non-SMTP message
2827 include only those that have been read before the error was detected.
2828 (In the case of an ACL rejection, they have all been read.)
2830 PH/69 The HTML version of the specification is now built in a directory called
2831 spec_html instead of spec.html, because the latter looks like a path with
2832 a MIME-type, and this confuses some software.
2834 PH/70 Catch two compiler warnings in sieve.c.
2836 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2837 function verify_get_ident() calls ip_connect() to connect a socket, but
2838 if the "connect()" function timed out, ip_connect() used to close the
2839 socket. However, verify_get_ident() also closes the socket later, and in
2840 between Exim writes to the log, which may get opened at this point. When
2841 the socket was closed in ip_connect(), the log could get the same file
2842 descriptor number as the socket. This naturally causes chaos. The fix is
2843 not to close the socket in ip_connect(); the socket should be closed by
2844 the function that creates it. There was only one place in the code where
2845 this was missing, in the iplookup router, which I don't think anybody now
2846 uses, but I've fixed it anyway.
2848 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2849 well as to direct DNS lookups. Otherwise the handling of names in host
2850 lists is inconsistent and therefore confusing.
2856 PH/01 Two changes to the default runtime configuration:
2858 (1) Move the checks for relay_from_hosts and authenticated clients from
2859 after to before the (commented out) DNS black list checks.
2861 (2) Add control=submission to the relay_from_hosts and authenticated
2862 clients checks, on the grounds that messages accepted by these
2863 statements are most likely to be submissions.
2865 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2867 (1) Generate an error if the third argument for the ${prvs expansion is
2870 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2873 (3) Reset the variables that are obtained from the first argument of
2874 ${prvscheck and used in the second argument before leaving the code,
2875 because their memory is reclaimed, so using them afterwards may do
2878 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2879 one (it's much easier than Tom thought :-).
2881 (5) Because of (4), we can now allow for the use of $prvscheck_result
2882 inside the third argument.
2884 PH/03 For some reason, the default setting of PATH when running a command from
2885 a pipe transport was just "/usr/bin". I have changed it to
2888 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2889 anything to be listed in the output from -bV.
2891 PH/05 When a filter generated an autoreply, the entire To: header line was
2892 quoted in the delivery log line, like this:
2894 => >A.N.Other <ano@some.domain> <original@ddress> ...
2896 This has been changed so that it extracts the operative address. There
2897 may be more than one such address. If so, they are comma-separated, like
2900 => >ano@some.domain,ona@other.domain <original@ddress> ...
2902 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2903 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2904 not being looked up in the rDNS to get a host name, Exim was showing the
2905 IP address twice in Received: lines, even though the IP addresses were
2906 identical. For example:
2908 Received: from [1.2.3.4] (helo=[1.2.3.4])
2910 However, if the real host name was known, it was omitting the HELO data
2911 if it matched the actual IP address. This has been tidied up so that it
2912 doesn't show the same IP address twice.
2914 PH/07 When both +timestamp and +memory debugging was on, the value given by
2915 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2916 called by the debug printing, thereby overwriting the timestamp buffer.
2917 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2919 PH/08 When the original message was included in an autoreply transport, it
2920 always said "this is a copy of the message, including all the headers",
2921 even if body_only or headers_only was set. It now gives an appropriate
2924 PH/09 Applied a patch from the Sieve maintainer which:
2926 o fixes some comments
2927 o adds the (disabled) notify extension core
2928 o adds some debug output for the result of if/elsif tests
2929 o points to the current vacation draft in the documentation
2930 and documents the missing references header update
2934 o fixes a bug in processing the envelope test (when testing
2935 multiple envelope elements, the last element determinted the
2938 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2939 Electronic Mail") by including:
2941 Auto-submitted: auto-generated
2943 in the messages that it generates (bounce messages and others, such as
2944 warnings). In the case of bounce messages for non-SMTP mesages, there was
2945 also a typo: it was using "Auto_submitted" (underscore instead of
2946 hyphen). Since every message generated by Exim is necessarily in response
2947 to another message, thes have all been changed to:
2949 Auto-Submitted: auto-replied
2951 in accordance with these statements in the RFC:
2953 The auto-replied keyword:
2955 - SHOULD be used on messages sent in direct response to another
2956 message by an automatic process,
2958 - MUST NOT be used on manually-generated messages,
2960 - MAY be used on Delivery Status Notifications (DSNs) and Message
2961 Disposition Notifications (MDNs),
2963 - MUST NOT be used on messages generated by automatic or periodic
2964 processes, except for messages which are automatic responses to
2967 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2968 to the default Received: header definition.
2970 PH/12 Added log selector acl_warn_skipped (default on).
2972 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2973 variables because (a) they are in the wrong storage pool and (b) even if
2974 they were copied, it wouldn't work properly because of the caching.
2976 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2977 checking when decoding. Apparently there are clients that generate
2978 overlong encoded strings. Why am I not surprised?
2980 PH/15 If the first argument of "${if match_address" was not empty, but did not
2981 contain an "@" character, Exim crashed. Now it writes a panic log message
2982 and treats the condition as false.
2984 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2986 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2987 extension "envelope-auth". The code is finished and in agreement with
2988 other implementations, but there is no documentation so far and in fact,
2989 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2990 not changing the active code.
2992 Print executed "if" and "elsif" statements when debugging is used. This
2993 helps a great deal to understand what a filter does.
2995 Document more things not specified clearly in RFC3028. I had all this
2996 sorted out, when out of a sudden new issues came to my mind. Oops."
2998 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3001 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3002 canonical form (as documented). However, after a host name lookup from
3003 the IP address, check_host() was doing a simple string comparison with
3004 addresses acquired from the DNS when checking that the found name did
3005 have the original IP as one of its addresses. Since any found IPv6
3006 addresses are likely to be in abbreviated form, the comparison could
3007 fail. Luckily, there already exists a function for doing the comparison
3008 by converting both addresses to binary, so now that is used instead of
3009 the text comparison.
3011 PH/20 There was another similar case to PH/19, when a complete host name was
3012 given in a host list; looking up its IP address could give an abbreviated
3013 form, whereas the current host's name might or might not be abbreviated.
3014 The same fix has been applied.
3020 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3021 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3024 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3025 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3027 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3029 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3030 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3031 Also, for "Auto-Submitted", treat anything other than "no" as
3032 non-personal, in accordance with RFC 3834. (Previously it treated
3033 anything starting "auto-" as non-personal.)
3035 TF/01 The control=submission/name=... option had a problem with syntax
3036 errors if the name included a slash character. The /name= option
3037 now slurps the rest of the string, so it can include any characters
3038 but it must come last in the list of options (after /sender_retain
3041 PH/05 Some modifications to the interface to the fake nameserver for the new
3049 TK/01 Added the "success_on_redirect" address verification option. See
3050 NewStuff for rationale and an example.
3052 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3054 PH/02 Patch to exigrep to allow it to work on syslog lines.
3056 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3057 fread() to skip over the body file's header line, because in Cygwin the
3058 header line is locked and is inaccessible.
3060 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3061 co-exist for some time) to make it clear that it is the Exim ID that is
3062 referenced, not the Message-ID: header line.
3064 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3065 string_format() function, because snprintf() does not exist on all
3068 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3069 ${stat: expansion item.
3071 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3072 protocol synchronization error", to keep the pedants happy.
3074 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3075 well as for IRIX systems, when gcc is being used. See the host.c source
3078 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3080 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3083 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3084 transport and to the smtp transport in LMTP mode.
3086 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3088 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3089 run a filter in a subprocess. This could lead to confusion in subsequent
3090 lookups in the parent process. There should also be a search_tidyup() at
3091 the end of the subprocess.
3093 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3094 only if the host matched helo_try_verify_hosts, which caused the
3095 verification to occur when the EHLO/HELO command was issued. The ACL just
3096 tested the remembered result. Now, if a previous verification attempt has
3097 not happened, "verify = helo" does it there and then.
3099 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3101 TK/03 Fix log output including CR from clamd.
3103 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3104 exist provoked a memory error which could cause a segfault.
3106 PH/15 Installed PCRE 6.2
3108 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3110 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3111 of the problem. Specifically, suggested +O2 rather than +O1 for the
3114 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3116 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3117 record, the port was not correctly logged when the outgoing_port log
3118 selector was set (it logged the transort's default port).
3120 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3121 fallback_hosts, and "hosts" in the smtp transport.
3123 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3124 host errors such as "Connection refused".
3126 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3127 authentication with radiusclient 0.4.9:
3129 - Error returned from rc_read_config was caught wrongly
3130 - Username/password not passed on to radius server due to wrong length.
3132 The presumption is that some radiusclient API changes for 4.51/PH/17
3133 were not taken care of correctly. The code is still untested by me (my
3134 Linux distribution still has 0.3.2 of radiusclient), but it was
3135 contributed by a Radius user.
3137 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3138 expanding the "port" option of the smtp transport.
3140 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3141 while reading a MIME header. Thanks to Tom Hughes for a patch.
3143 PH/24 Include config.h inside local_scan.h so that configuration settings are
3146 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3147 that in an ACL for RCPT (for example), you can examine exactly what was
3150 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3151 commands, but it was not correctly comparing the address with the actual
3152 client host address. Thus, it would show the EHLO address in Received:
3153 header lines when this was not necessary.
3155 PH/27 Added the % operator to ${eval:}.
3157 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3158 it should be ignoring failures (because with -C, for example, it has lost
3159 privilege). It wasn't ignoring creation failures other than "already
3162 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3165 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3166 because an input line was too long, either on its own, or by virtue of
3167 too many continuations, the temporary file was not being removed, and the
3168 return code was incorrect.
3170 PH/31 Missing "BOOL" in function definition in filtertest.c.
3172 PH/32 Applied Sieve patches from the maintainer.
3174 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3176 PH/33 Added "verify = not_blind".
3178 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3179 Local/Makefile (with some defaults set). These are used in built scripts
3180 such as exicyclog, but they have never been used in the exim_install
3181 script (though there are many overriding facilities there). I have
3182 arranged that the exim_install script now takes note of these two
3185 PH/35 Installed configuration files for Dragonfly.
3187 PH/36 When a locally submitted message by a trusted user did not contain a
3188 From: header, and the sender address was obtained from -f or from an SMTP
3189 MAIL command, and the trusted user did not use -F to supply a sender
3190 name, $originator_name was incorrectly used when constructing a From:
3191 header. Furthermore, $originator_name was used for submission mode
3192 messages from external hosts without From: headers in a similar way,
3193 which is clearly wrong.
3195 PH/37 Added control=suppress_local_fixups.
3197 PH/38 When log_selector = +received_sender was set, and the addition of the
3198 sender made the log line's construction buffer exactly full, or one byte
3199 less than full, an overflow happened when the terminating "\n" was
3202 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3203 when the result of a list match is failure because a DNS lookup failed.
3205 PH/40 RM_COMMAND is now used in the building process.
3207 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3208 the "build-* directories that it finds.
3210 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3211 address in a domain literal was a prefix of an interface address.
3213 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3214 when verifying a sender address, unless rewrite_headers is false.
3216 PH/44 Wrote a long comment about why errors_to addresses are verified as
3217 recipients, not senders.
3219 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3220 the ratelimit ACL was added.
3222 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3224 PH/46 Added extra information about PostgreSQL errors to the error string.
3226 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3227 avoiding the need to install special zones in a real server. This is
3228 backwards compatible; if it can't find the fake resolver, it drops back.
3229 Thus, both old and new test suites can be run.
3231 TF/02 Added util/ratelimit.pl
3233 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3234 clock is set back in time.
3236 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3237 Candler <B.Candler@pobox.com>.
3239 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3240 OK for addresses that are the result of redirections.
3242 PH/48 A number of further additions for the benefit of the new test suite,
3243 including a fake gethostbyname() that interfaces to the fake DNS resolver
3246 TF/06 The fix for widen_domains has also been applied to qualify_single and
3247 search_parents which are the other dnslookup options that can cause
3250 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3253 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3255 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3256 service" if it encountered an MX record with an empty target hostname.
3257 The message is now "an MX or SRV record indicated no SMTP service".
3259 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3260 if the DNS of the sending site is misconfigured. This is quite a
3261 common situation. This change restores the behaviour of treating a
3262 helo verification defer as a failure.
3264 PH/51 If self=fail was set on a router, the bounce message did not include the
3265 actual error message.
3271 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3273 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3274 command itself ended in error, the underlying message about the transport
3275 filter timeout was being overwritten with the pipe command error. Now the
3276 underlying error message should be appended to the second error message.
3278 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3280 PH/02 Reduce the amount of output that "make" produces by default. Full output
3281 can still be requested.
3283 PH/03 The warning log line about a condition test deferring for a "warn" verb
3284 was being output only once per connection, rather than after each
3285 occurrence (because it was using the same function as for successful
3286 "warn" verbs). This seems wrong, so I have changed it.
3288 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3289 it should not have, which might have caused a crash in the right
3290 circumstances, but probably never did.
3292 PH/04 Installed a modified version of Tony Finch's patch to make submission
3293 mode fix the return path as well as the Sender: header line, and to
3294 add a /name= option so that you can make the user's friendly name appear
3297 TF/03 Added the control = fakedefer ACL modifier.
3299 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3300 Mark Lowes for thorough testing.
3302 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3304 TK/03 Merged latest SRS patch from Miles Wilton.
3306 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3307 with the definition in sysexits.h (which is #included earlier).
3308 Fortunately, Exim does not actually use EX_OK. The code used to try to
3309 preserve the sysexits.h value, by assumimg that macro definitions were
3310 scanned for macro replacements. I have been disabused of this notion,
3311 so now the code just undefines EX_OK before #including unistd.h.
3313 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3314 in the smtp transport. When a block could not be written in a single
3315 write() function, the timeout was being re-applied to each part-write.
3316 This seems wrong - if the receiver was accepting one byte at a time it
3317 would take for ever. The timeout is now adjusted when this happens. It
3318 doesn't have to be particularly precise.
3320 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3321 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3323 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3324 without a domain if the check to <postmaster@domain> fails.
3326 SC/01 Eximstats: added -xls and the ability to specify output files
3327 (patch written by Frank Heydlauf).
3329 SC/02 Eximstats: use FileHandles for outputing results.
3331 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3333 SC/04 Eximstats: fixed display of large numbers with -nvr option
3335 SC/05 Eximstats: fixed merging of reports with empty tables.
3337 SC/06 Eximstats: added the -include_original_destination flag
3339 SC/07 Eximstats: removed tabs and trailing whitespace.
3341 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3343 TK/06 MBOX spool code: Add real "From " MBOX separator line
3344 so the .eml file is really in mbox format (even though
3345 most programs do not really care). Patch from Alex Miller.
3347 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3348 The latter is generated from $received_to and is only set if the
3349 message has one envelope recipient. SA can use these headers,
3350 obviously out-of-the-box. Patch from Alex Miller.
3352 PH/08 The ${def test on a variable was returning false if the variable's
3353 value was "0", contrary to what the specification has always said!
3354 The result should be true unless the variable is empty.
3356 PH/09 The syntax error of a character other than { following "${if
3357 def:variable_name" (after optional whitespace) was not being diagnosed.
3358 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3359 accidental colon was present, for example, could give incorrect results.
3361 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3362 result is used (not including appendfile, where other changes are about
3365 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3366 This involved changing a lot of size variables from int to off_t. It
3367 should work with maildirs and everything.
3369 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3370 spamd dying while we are connected to it.
3372 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3375 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3376 function for BDB 4.3.
3378 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3380 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3381 the sources for PCRE have been re-arranged and more files are now
3384 PH/15 The code I had for printing potentially long long variables in PH/11
3385 above was not the best (it lost precision). The length of off_t variables
3386 is now inspected at build time, and an appropriate printing format (%ld
3387 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3388 to be "long long int" or "long int". This is needed for the internal
3389 formatting function string_vformat().
3391 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3392 the configuration file to be ":syslog", then the script "guesses" where
3393 the logs files are, rather than using the compiled in default. In our
3394 case the guess is not the same as the compiled default, so the script
3395 suddenly stopped working when I started to use syslog. The patch checks
3396 to see if log_file_path is "". If so, it attempts to read it from exim
3397 with no configuration file to get the compiled in version, before it
3398 falls back to the previous guessing code."
3400 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3401 implementing BATV in an Exim configuration. See NewStuff for the gory
3404 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3405 Makefile that are specific to HP-UX.
3407 PH/18 If the "use_postmaster" option was set for a recipient callout together
3408 with the "random" option, the postmaster address was used as the MAIL
3409 FROM address for the random test, but not for the subsequent recipient
3410 test. It is now used for both.
3412 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3413 patch removes a few documentation additions to RFC 3028, because the
3414 latest draft now contains them. It adds the new en;ascii-case comparator
3415 and a new error check for 8bit text in MIME parts. Comparator and
3416 require names are now matched exactly. I enabled the subaddress
3417 extension, but it is not well tested yet (read: it works for me)."
3419 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3420 rework some of the code of TK/09 above to avoid the hardwired use of
3421 "%lld" and "long long". Replaced the call to snprintf() with a call to
3424 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3425 records point to non-existent hosts", "retry timeout exceeded", and
3426 "retry time not reached for any host after a long failure period".
3428 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3429 experimental DomainKeys support:
3431 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3432 (2) On an error such as an illegally used "control", the wrong name for
3433 the control was given.
3435 These problems did NOT occur unless DomainKeys support was compiled.
3437 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3439 PH/24 Added ${if match_ip condition.
3441 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3442 databases so that it will be absolutely obvious if a crash occurs in the
3443 DB library. This is a regular occurrence (often caused by mis-matched
3446 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3447 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3448 picked up on a user's system that detects such things. There doesn't seem
3449 to be a gcc warning option for this - only an attribute that has to be
3450 put on the function's prototype. It seems that in Fedora Core 4 they have
3451 set this on a number of new functions. No doubt there will be more in due
3454 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3455 specify a transport. However, if an address that was verified by such a
3456 router was the subject of a callout, Exim crashed because it tried to
3457 read the rcpt_include_affixes from the non-existent transport. Now it
3458 just assumes that the setting of that option is false. This bug was
3459 introduced by 4.51/PH/31.
3461 PH/28 Changed -d+all to exclude +memory, because that information is very
3462 rarely of interest, but it makes the output a lot bigger. People tend to
3463 do -d+all out of habit.
3465 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3466 code in os-type was giving problems when libc.so lives in lib64, like on
3469 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3470 aren't the modern standard, and it seems that some systems' include files
3471 don't always have them. Exim was already checking for some of the newer
3472 ones like T_AAAA, and defining it itself. I've added checks for all the
3473 record types that Exim uses.
3475 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3476 not automatically generating a new one, as it is supposed to. This
3477 prevented TLS from working. If the file did exist, but contained invalid
3478 data, a new version was generated, as expected. It was only the case of a
3479 non-existent file that was broken.
3481 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3482 with a change in libdomainkeys > 0.64.
3484 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3485 from DNS. If the selector record carries the flag, it now has
3486 precedence over the domain-wide flag.
3488 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3490 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3491 the use of a port name that isn't defined in /etc/services) occurred, the
3492 message was deferred as in a normal delivery, and thus remained on the
3493 spool, instead of being failed because of the mua_wrapper setting. This
3494 is now fixed, and I tidied up some of the mua_wrapper messages at the
3497 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3498 the messages in a hash of arrays rather than using individual hashes.
3499 This is a bit cleaner and results in dramatic memory savings, albeit
3500 at a slight CPU cost.
3502 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3503 as requested by Marc Sherman.
3505 SC/10 Eximstats: added histograms for user specified patterns as requested
3508 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3510 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3511 fopen() in the content-scanning modules that did not already have it.
3517 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3518 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3520 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3522 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3524 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3525 to test Sieve filters that use "vacation".
3527 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3528 that changes the way the GnuTLS parameters are stored in the cache file.
3529 The new format can be generated externally. For backward compatibility,
3530 if the data in the cache doesn't make sense, Exim assumes it has read an
3531 old-format file, and it generates new data and writes a new file. This
3532 means that you can't go back to an older release without removing the
3535 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3536 work if there are any delivery delays because "one_time" forces the
3537 parent to be marked "delivered", so its unseen clone is never tried
3538 again. For this reason, Exim now forbids the simultaneous setting of
3541 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3542 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3543 entry if you want to know the details. The fix, however, neglected to
3544 consider the case when local delivery batching is involved. The test for
3545 "previously delivered" was not happening when checking to see if an
3546 address could be batched with a previous (undelivered) one; under
3547 certain circumstances this could lead to multiple deliveries to the same
3550 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3551 in its include files, and this causes problems building Exim.
3553 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3554 header_syntax) but Exim was just ignoring anything given after a slash.
3555 In particular, this caused confusion with an attempt to use "verify =
3556 reverse_host_lookup/defer_ok". An error is now given when options are
3557 supplied for verify items that do not have them. (Maybe reverse_host_
3558 lookup should have a defer_ok option, but that's a different point.)
3560 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3561 defined by RFC 821) to 2048, because there were problems with some AUTH
3562 commands, and RFC 1869 says the size should be increased for extended
3563 SMTP commands that take arguments.
3565 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3568 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3569 "unknown" error; now it says that the functionality isn't in the binary.
3571 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3572 an address' error message when a string expansion fails (syntax or
3573 whatever). Otherwise the password may appear in the log. Following change
3574 PH/42 below, there is no longer a chance of it appearing in a bounce
3577 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3579 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3580 of its addresses was ever tried. (Bugzilla bug #2.)
3582 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3583 the result incorrectly in the debug output. (It correctly added a newline
3584 to what was transported.)
3586 TF/01 Added $received_time.
3588 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3589 commented out examples of how to interface to a virus scanner and to
3590 SpamAssassin. Also added commented examples of av_scanner and
3591 spamd_address settings.
3593 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3594 and controls are allowed in which ACLs. There were a couple of minor
3595 errors. Some of the entries in the conditions table (which is a table of
3596 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3597 negation of where the condition IS allowed.
3599 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3601 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3602 header file does not have a version number, so I've had to invent a new
3603 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3604 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3605 radiusclient), but it was contributed by a Radius user.
3607 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3608 files or directories, for OpenSSL.
3610 PH/19 When an Exim process that is running as root has to create an Exim log
3611 file, it does so in a subprocess that runs as exim:exim so as to get the
3612 ownership right at creation (otherwise, other Exim processes might see
3613 the file with the wrong ownership). There was no test for failure of this
3614 fork() call, which would lead to the process getting stuck as it waited
3615 for a non-existent subprocess. Forks do occasionally fail when resources
3616 run out. I reviewed all the other calls to fork(); they all seem to check
3619 PH/20 When checking for unexpected SMTP input at connect time (before writing
3620 the banner), Exim was not dealing correctly with a non-positive return
3621 from the read() function. If the client had disconnected by this time,
3622 the result was a log entry for a synchronization error with an empty
3623 string after "input=" when read() returned zero. If read() returned -1
3624 (an event I could not check), uninitialized data bytes were printed.
3625 There were reports of junk text (parts of files, etc) appearing after
3628 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3630 PH/22 Added support for macro redefinition, and (re)definition in between
3631 driver and ACL definitions.
3633 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3634 forgetting to use the resulting value; it was using the unexpanded value.
3636 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3637 hadn't been configured. The fix is from Juergen Kreileder, who
3638 understands it better than I do:
3640 "Here's what I see happening with three configured cyrus_sasl
3641 authenticators configured (plain, login, cram-md5):
3643 On startup auth_cyrus_sasl_init() gets called for each of these.
3644 This means three calls to sasl_listmech() without a specified mech_list.
3645 => SASL tests which mechs of all available mechs actually work
3646 => three warnings about OTP not working
3647 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3649 With the patch, sasl_listmech() also gets called three times. But now
3650 SASL's mech_list option is set to the server_mech specified in the the
3651 authenticator. Or in other words, the answer from sasl_listmech()
3652 gets limited to just the mech you're testing for (which is different
3654 => the return list contains just 'plain' or 'login', 'cram-md5' or
3655 nothing depending on the value of ob->server_mech.
3657 I've just tested the patch: Authentication still works fine,
3658 unavailable mechs specified in the exim configuration are still
3659 caught, and the auth.log warnings about OTP are gone."
3661 PH/25 When debugging is enabled, the contents of the command line are added
3662 to the debugging output, even when log_selector=+arguments is not
3665 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3666 answer is "GNU", and only if the return is "GNU/something" is the answer
3669 PH/27 $acl_verify_message is now set immediately after the failure of a
3670 verification in an ACL, and so is available in subsequent modifiers. In
3671 particular, the message can be preserved by coding like this:
3673 warn !verify = sender
3674 set acl_m0 = $acl_verify_message
3676 Previously, $acl_verify_message was set only while expanding "message"
3677 and "log_message" when a very denied access.
3679 PH/28 Modified OS/os.c-Linux with
3681 -#ifndef OS_LOAD_AVERAGE
3682 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3684 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3685 nomenclature these days.)
3687 PH/29 Installed patch from the Sieve maintainer that adds the options
3688 sieve_useraddress and sieve_subaddress to the redirect router.
3690 PH/30 In these circumstances:
3691 . Two addresses routed to the same list of hosts;
3692 . First host does not offer TLS;
3693 . First host accepts first address;
3694 . First host gives temporary error to second address;
3695 . Second host offers TLS and a TLS session is established;
3696 . Second host accepts second address.
3697 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3698 and peerdn, if requested) that were in fact used only for the second
3701 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3702 attention to any local part prefix or suffix that was matched by the
3703 router that accepted the address. It now behaves in the same way as it
3704 does for delivery: the affixes are removed from the local part unless
3705 rcpt_include_affixes is set on the transport.
3707 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3708 timeout during the DATA phase of an incoming message.
3710 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3711 applied a patch sanctioned by the Sieve maintainer.
3713 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3714 the uid or gid is negative. A case of a negative gid caused this to be
3715 noticed. The fix allows for either to be negative.
3717 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3718 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3721 PH/36 The change PH/12 above was broken. Fixed it.
3723 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3724 the grounds that routing the same address twice would always produce the
3725 same answer. This might have been true once, but it is certainly no
3726 longer true now. Routing a child address may depend on the previous
3727 routing that produced that child. Some complicated redirection strategies
3728 went wrong when messages had multiple recipients, and made Exim's
3729 behaviour dependent on the order in which the addresses were given.
3731 I have moved the duplicate checking until after the routing is complete.
3732 Exim scans the addresses that are assigned to local and remote
3733 transports, and removes any duplicates. This means that more work will be
3734 done, as duplicates will always all be routed, but duplicates are
3735 presumably rare, so I don't expect this is of any significance.
3737 For deliveries to pipes, files, and autoreplies, the duplicate checking
3738 still happens during the routing process, since they are not going to be
3741 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3742 It corrects a timeout issue with spamd. This is Ian's comment: "The
3743 background is that sometimes spamd either never reads data from a
3744 connection it has accepted, or it never writes response data. The exiscan
3745 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3746 blindly assumes that writes won't block so it may never time out."
3748 PH/39 Allow G after quota size as well as K and M.
3750 PH/40 The value set for $authenticated_id in an authenticator may not contain
3751 binary zeroes or newlines because the value is written to log lines and
3752 to spool files. There was no check on this. Now the value is run through
3753 the string_printing() function so that such characters are converted to
3754 printable escape sequences.
3756 PH/41 $message_linecount is a new variable that contains the total number of
3757 lines in the message. Compare $body_linecount, which is the count for the
3760 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3761 bounce and delay warning messages, except in certain special cases, which
3764 (a) An SMTP error message from a remote host;
3765 (b) A message specified in a :fail: redirection;
3766 (c) A message specified in a "fail" command in a system filter;
3767 (d) A message specified in a FAIL return from the queryprogram router;
3768 (e) A message specified by the cannot_route_message router option.
3770 In these cases only, Exim does include the error details in bounce and
3771 warning messages. There are also a few cases where bland messages such
3772 as "unrouteable address" or "local delivery error" are given.
3774 PH/43 $value is now also set for the "else" part of a ${run expansion.
3776 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3777 being worked on, but at least Exim now implements the latest version to
3780 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3781 process to complete was treated as a delivery failure, a timeout while
3782 writing the message to the pipe was logged, but erroneously treated as a
3783 successful delivery. Such timeouts include transport filter timeouts. For
3784 consistency with the overall process timeout, these timeouts are now
3785 treated as errors, giving rise to delivery failures by default. However,
3786 there is now a new Boolean option for the pipe transport called
3787 timeout_defer, which, if set TRUE, converts the failures into defers for
3788 both kinds of timeout. A transport filter timeout is now identified in
3791 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3792 systems where "make" and "gmake" are different, calling "gmake" at top
3793 level broke things. I've arranged for the value of $(MAKE) to be passed
3794 from the Makefile to this script so that it can call the same version of
3798 A note about Exim versions 4.44 and 4.50
3799 ----------------------------------------
3801 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3802 changes of various kinds. As a consequence, a big documentation update was
3803 needed. This delayed the release for rather longer than seemed good, especially
3804 in the light of a couple of (minor) security issues. Therefore, the changes
3805 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3806 release. So 4.44 and 4.50 are in effect two different branches that both start
3809 I have left the 4.50 change log unchanged; it contains all the changes since
3810 4.43. The change log for 4.44 is below; many of its items are identical to
3811 those for 4.50. This seems to be the most sensible way to preserve the
3812 historical information.
3818 1. Minor wording change to the doc/README.SIEVE file.
3820 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3821 computation of the current number of files was incorrect.
3823 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3824 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3827 4. Give more explanation in the error message when the command for a transport
3828 filter fails to execute.
3830 5. There are several places where Exim runs a non-Exim command in a
3831 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3832 was being done only for the command run by the queryprogram router. It is
3833 now done for all such subprocesses. The other cases are: ${run, transport
3834 filters, and the commands run by the lmtp and pipe transports.
3836 6. Added CONFIGURE_GROUP build-time option.
3838 7. Some older OS have a limit of 256 on the maximum number of file
3839 descriptors. Exim was using setrlimit() to set 1000 as a large value
3840 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3841 systems. I've change it so that if it can't get 1000, it tries for 256.
3843 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3844 was an oversight, and furthermore, ever since the addition of extra
3845 controls (e.g. 4.43/32), the checks on when to allow different forms of
3846 "control" were broken. There should now be diagnostics for all cases when a
3847 control that does not make sense is encountered.
3849 9. Added the /retain_sender option to "control=submission".
3851 10. $recipients is now available in the predata ACL (oversight).
3853 11. Tidy the search cache before the fork to do a delivery from a message
3854 received from the command line. Otherwise the child will trigger a lookup
3855 failure and thereby defer the delivery if it tries to use (for example) a
3856 cached ldap connection that the parent has called unbind on.
3858 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3859 of $address_data from the recipient verification was clobbered by the
3860 sender verification.
3862 13. The value of address_data from a sender verification is now available in
3863 $sender_address_data in subsequent conditions in the ACL statement.
3865 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3867 15. Added a new option "connect=<time>" to callout options, to set a different
3870 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3871 was its contents. (It was OK if the option was not defined at all.)
3873 17. A "Completed" log line is now written for messages that are removed from
3874 the spool by the -Mrm option.
3876 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3877 information about exactly what failed.
3879 19. Added -dd to debug only the daemon process.
3881 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3882 handles timeouts, both on the server side and network timeouts. Renamed the
3883 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3885 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3886 It is now set to "smtps".
3888 22. $host_address is now set to the target address during the checking of
3889 ignore_target_hosts.
3891 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3892 being passed; this would have caused $sender_host_name to have been used if
3893 matching the list had actually called for a host name (not very likely,
3894 since this list is usually IP addresses). A host name is now passed as
3897 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3898 code that specifies a non-restarting handler (typically sigaction() in
3899 modern systems) in an attempt to fix a rare and obscure crash bug.
3901 25. Narrowed the window for a race in the daemon that could cause it to ignore
3902 SIGCHLD signals. This is not a major problem, because they are used only to
3903 wake it up if nothing else does.
3905 26. A malformed maildirsize file could cause Exim to calculate negative values
3906 for the mailbox size or file count. Odd effects could occur as a result.
3907 The maildirsize information is now recalculated if the size or filecount
3910 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3911 support for a long time. Removed HAVE_SYS_VFS_H.
3913 28. Installed the latest version of exipick from John Jetmore.
3915 29. In an address list, if the pattern was not a regular expression, an empty
3916 subject address (from a bounce message) matched only if the pattern was an
3917 empty string. Non-empty patterns were not even tested. This was the wrong
3918 because it is perfectly reasonable to use an empty address as part of a
3919 database query. An empty address is now tested by patterns that are
3920 lookups. However, all the other forms of pattern expect the subject to
3921 contain a local part and a domain, and therefore, for them, an empty
3922 address still always fails if the pattern is not itself empty.
3924 30. Exim went into a mad DNS loop when attempting to do a callout where the
3925 host was specified on an smtp transport, and looking it up yielded more
3926 than one IP address.
3928 31. Re-factored the code for checking spool and log partition space into a
3929 function that finds that data and another that does the check. The former
3930 is then used to implement four new variables: $spool_space, $log_space,
3931 $spool_inodes, and $log_inodes.
3933 32. The RFC2047 encoding function was originally intended for short strings
3934 such as real names; it was not keeping to the 75-character limit for
3935 encoded words that the RFC imposes. It now respects the limit, and
3936 generates multiple encoded words if necessary. To be on the safe side, I
3937 have increased the buffer size for the ${rfc2047: expansion operator from
3940 33. It is now permitted to omit both strings after an "if" condition; if the
3941 condition is true, the result is "true". As before, when the second string
3942 is omitted, a false condition yields an empty string. This makes it less
3943 cumbersome to write custom ACL and router conditions.
3945 34. Failure to deliver a bounce message always caused it to be frozen, even if
3946 there was an errors_to setting on the router. The errors_to setting is now
3949 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3950 canonical form (fully expanded) before being placed in
3951 $sender_host_address.
3953 36. The table in the code that translates DNS record types into text (T_A to
3954 "A" for instance) was missing entries for NS and CNAME. It is just possible
3955 that this could have caused confusion if both these types were looked up
3956 for the same domain, because the text type is used as part of Exim's
3957 per-process caching. But the chance of anyone hitting this buglet seems
3960 37. The dnsdb lookup has been extended in a number of ways.
3962 (1) There is a new type, "zns", which walks up the domain tree until it
3963 finds some nameserver records. It should be used with care.
3965 (2) There is a new type, "mxh", which is like "mx" except that it returns
3966 just the host names, not the priorities.
3968 (3) It is now possible to give a list of domains (or IP addresses) to be
3969 looked up. The behaviour when one of the lookups defers can be
3970 controlled by a keyword.
3972 (4) It is now possible to specify the separator character for use when
3973 multiple records are returned.
3975 38. The dnslists ACL condition has been extended: it is now possible to supply
3976 a list of IP addresses and/or domains to be looked up in a particular DNS
3979 39. Added log_selector=+queue_time_overall.
3981 40. When running the queue in the test harness, wait just a tad after forking a
3982 delivery process, to get repeatability of debugging output.
3984 41. Include certificate and key file names in error message when GnuTLS fails
3985 to set them up, because the GnuTLS error message doesn't include the name
3986 of the failing file when there is a problem reading it.
3988 42. Allow both -bf and -bF in the same test run.
3990 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3992 44. The "Exiscan patch" is now merged into the mainline Exim source.
3994 45. Sometimes the final signoff response after QUIT could fail to get
3995 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3996 < 0 before doing a fflush(). This bug looks as though it goes back to the
3997 introduction of TLS in release 3.20, but "sometimes" must have been rare
3998 because the tests only now provoked it.
4000 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4001 (this can affect the format of dates).
4003 47. exim_tidydb, when checking for the continued existence of a message for
4004 which it has found a message-specific retry record, was not finding
4005 messages that were in split spool directories. Consequently, it was
4006 deleting retry records that should have stayed in existence.
4008 48. Steve fixed some bugs in eximstats.
4010 49. The SPA authentication driver was not abandoning authentication and moving
4011 on to the next authenticator when an expansion was forced to fail,
4012 contradicting the general specification for all authenticators. Instead it
4013 was generating a temporary error. It now behaves as specified.
4015 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4016 (the order specifies the preference for clients). The order is now AES256,
4017 AES128, 3DES, ARCFOUR128.
4019 51. Small patch to Sieve code - explicitly set From: when generating an
4022 52. Exim crashed if a remote delivery caused a very long error message to be
4023 recorded - for instance if somebody sent an entire SpamAssassin report back
4024 as a large number of 550 error lines. This bug was coincidentally fixed by
4025 increasing the size of one of Exim's internal buffers (big_buffer) that
4026 happened as part of the Exiscan merge. However, to be on the safe side, I
4027 have made the code more robust (and fixed the comments that describe what
4030 53. Now that there can be additional text after "Completed" in log lines (if
4031 the queue_time_overall log selector is set), a one-byte patch to exigrep
4032 was needed to allow it to recognize "Completed" as not the last thing in
4035 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4036 patch that reportedly fixes this has been added. I am not expert enough to
4037 create a test for it. This is what the patch creator wrote:
4039 "I found a little strange behaviour of ldap code when working with
4040 Windows 2003 AD Domain, where users was placed in more than one
4041 Organization Units. When I tried to give exim partial DN, the exit code
4042 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4043 But simultaneously result of request was absolutely normal ldap result,
4044 so I produce this patch..."
4046 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4047 REFERENCE, so I have modified the code to exclude the patch when that macro
4050 55. Some experimental protocols are using DNS PTR records for new purposes. The
4051 keys for these records are domain names, not reversed IP addresses. The
4052 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4053 leaves it alone. Component reversal etc. now happens only for IP addresses.
4056 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4058 57. Double the size of the debug message buffer (to 2048) so that more of very
4059 long debug lines gets shown.
4061 58. The exicyclog utility now does better if the number of log files to keep
4062 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4064 59. Two changes related to the smtp_active_hostname option:
4066 (1) $smtp_active_hostname is now available as a variable.
4067 (2) The default for smtp_banner uses $smtp_active_hostname instead
4068 of $primary_hostname.
4070 60. The host_aton() function is supposed to be passed a string that is known
4071 to be a valid IP address. However, in the case of IPv6 addresses, it was
4072 not checking this. This is a hostage to fortune. Exim now panics and dies
4073 if the condition is not met. A case was found where this could be provoked
4074 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4075 components; fortuitously, this particular loophole had already been fixed
4076 by change 4.50/55 above.
4078 If there are any other similar loopholes, the new check in host_aton()
4079 itself should stop them being exploited. The report I received stated that
4080 data on the command line could provoke the exploit when Exim was running as
4081 exim, but did not say which command line option was involved. All I could
4082 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4083 running as the user.
4086 61. There was a buffer overflow vulnerability in the SPA authentication code
4087 (which came originally from the Samba project). I have added a test to the
4088 spa_base64_to_bits() function which I hope fixes it.
4091 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4092 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4094 63. The daemon start-up calls getloadavg() while still root for those OS that
4095 need the first call to be done as root, but it missed one case: when
4096 deliver_queue_load_max is set with deliver_drop_privilege. This is
4097 necessary for the benefit of the queue runner, because there is no re-exec
4098 when deliver_drop_privilege is set.
4100 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4101 This has been fixed.
4103 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4104 was in use, was not putting the data itself into the right store pool;
4105 consequently, it could be overwritten for a subsequent message in the same
4106 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4109 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4111 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4112 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4113 the function were treating the return as a boolean value, which happened to
4114 work because 0=false and not-0=true, but is not correct code.
4116 68. The host_aton() function was not handling scoped IPv6 addresses (those
4117 with, for example, "%eth0" on the end) correctly.
4119 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4120 negated items (that is, ~something) in unsigned ints. Some compilers
4121 apparently mutter when there is no cast.
4123 70. If an address verification called from an ACL failed, and did not produce a
4124 user-specific message (i.e. there was only a "system" message), nothing was
4125 put in $acl_verify_message. In this situation, it now puts the system
4128 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4129 session; change 4.31/43 added the unwanted input to the log line - except
4130 that it did not do this in the start of session case. It now does.
4132 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4133 This is wrong and can cause the other end to generate a synchronization
4134 error if it is another Exim or anything else that does the synchronization
4135 check. A QUIT command is no longer sent after a timeout.
4137 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4138 during host lookups.
4140 74. The defer_ok option of callout verification was not working if it was used
4141 when verifying addresses in header lines, that is, for this case:
4143 verify = header_sender/callout=defer_ok
4145 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4146 those file descriptors could be used for SMTP connections. If anything
4147 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4148 could be sent to the SMTP client, causing chaos. The daemon now opens
4149 stdin, stdout, and stderr to /dev/null when it puts itself into the
4152 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4153 log by default. The user can override this with suitable Perl magic.
4155 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4156 the log message when discard triggers, was not working for the DATA ACL or
4157 for the non-SMTP ACL.
4159 78. Error message wording change in sieve.c.
4161 79. If smtp_accept_max_per_host was set, the number of connections could be
4162 restricted to fewer than expected, because the daemon was trying to set up
4163 a new connection before checking whether the processes handling previous
4164 connections had finished. The check for completed processes is now done
4165 earlier. On busy systems, this bug wouldn't be noticed because something
4166 else would have woken the daemon, and it would have reaped the completed
4169 80. If a message was submitted locally by a user whose login name contained one
4170 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4171 It caused a spool format error. I have fixed the spool reading code. A
4172 related problem was that the "from" clause in the Received: line became
4173 illegal because of the space(s). It is now covered by ${quote_local_part.
4175 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4178 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4180 83. Patch from Sieve maintainer to fix unterminated string problem in
4181 "vacation" handling.
4183 84. Some minor changes to the Linux configuration files to help with other
4184 OS variants using glibc.
4186 85. One more patch for Sieve to update vacation handling to latest spec.
4189 ----------------------------------------------------
4190 See the note above about the 4.44 and 4.50 releases.
4191 ----------------------------------------------------
4197 1. Change 4.43/35 introduced a bug that caused file counts to be
4198 incorrectly computed when quota_filecount was set in an appendfile
4201 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4202 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4205 3. Give more explanation in the error message when the command for a transport
4206 filter fails to execute.
4208 4. There are several places where Exim runs a non-Exim command in a
4209 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4210 was being done only for the command run by the queryprogram router. It is
4211 now done for all such subprocesses. The other cases are: ${run, transport
4212 filters, and the commands run by the lmtp and pipe transports.
4214 5. Some older OS have a limit of 256 on the maximum number of file
4215 descriptors. Exim was using setrlimit() to set 1000 as a large value
4216 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4217 systems. I've change it so that if it can't get 1000, it tries for 256.
4219 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4220 was an oversight, and furthermore, ever since the addition of extra
4221 controls (e.g. 4.43/32), the checks on when to allow different forms of
4222 "control" were broken. There should now be diagnostics for all cases when a
4223 control that does not make sense is encountered.
4225 7. $recipients is now available in the predata ACL (oversight).
4227 8. Tidy the search cache before the fork to do a delivery from a message
4228 received from the command line. Otherwise the child will trigger a lookup
4229 failure and thereby defer the delivery if it tries to use (for example) a
4230 cached ldap connection that the parent has called unbind on.
4232 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4233 of $address_data from the recipient verification was clobbered by the
4234 sender verification.
4236 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4237 was its contents. (It was OK if the option was not defined at all.)
4239 11. A "Completed" log line is now written for messages that are removed from
4240 the spool by the -Mrm option.
4242 12. $host_address is now set to the target address during the checking of
4243 ignore_target_hosts.
4245 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4246 being passed; this would have caused $sender_host_name to have been used if
4247 matching the list had actually called for a host name (not very likely,
4248 since this list is usually IP addresses). A host name is now passed as
4251 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4252 code that specifies a non-restarting handler (typically sigaction() in
4253 modern systems) in an attempt to fix a rare and obscure crash bug.
4255 15. Narrowed the window for a race in the daemon that could cause it to ignore
4256 SIGCHLD signals. This is not a major problem, because they are used only to
4257 wake it up if nothing else does.
4259 16. A malformed maildirsize file could cause Exim to calculate negative values
4260 for the mailbox size or file count. Odd effects could occur as a result.
4261 The maildirsize information is now recalculated if the size or filecount
4264 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4265 support for a long time. Removed HAVE_SYS_VFS_H.
4267 18. Updated exipick to current release from John Jetmore.
4269 19. Allow an empty sender to be matched against a lookup in an address list.
4270 Previously the only cases considered were a regular expression, or an
4273 20. Exim went into a mad DNS lookup loop when doing a callout where the
4274 host was specified on the transport, if the DNS lookup yielded more than
4277 21. The RFC2047 encoding function was originally intended for short strings
4278 such as real names; it was not keeping to the 75-character limit for
4279 encoded words that the RFC imposes. It now respects the limit, and
4280 generates multiple encoded words if necessary. To be on the safe side, I
4281 have increased the buffer size for the ${rfc2047: expansion operator from
4284 22. Failure to deliver a bounce message always caused it to be frozen, even if
4285 there was an errors_to setting on the router. The errors_to setting is now
4288 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4289 canonical form (fully expanded) before being placed in
4290 $sender_host_address.
4292 24. Updated eximstats to version 1.33
4294 25. Include certificate and key file names in error message when GnuTLS fails
4295 to set them up, because the GnuTLS error message doesn't include the name
4296 of the failing file when there is a problem reading it.
4298 26. Expand error message when OpenSSL has problems setting up cert/key files.
4301 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4302 (this can affect the format of dates).
4304 28. exim_tidydb, when checking for the continued existence of a message for
4305 which it has found a message-specific retry record, was not finding
4306 messages that were in split spool directories. Consequently, it was
4307 deleting retry records that should have stayed in existence.
4309 29. eximstats updated to version 1.35
4310 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4311 1.35 - bugfix such that pie charts by volume are generated correctly
4313 30. The SPA authentication driver was not abandoning authentication and moving
4314 on to the next authenticator when an expansion was forced to fail,
4315 contradicting the general specification for all authenticators. Instead it
4316 was generating a temporary error. It now behaves as specified.
4318 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4319 (the order specifies the preference for clients). The order is now AES256,
4320 AES128, 3DES, ARCFOUR128.
4322 31. Small patch to Sieve code - explicitly set From: when generating an
4325 32. Exim crashed if a remote delivery caused a very long error message to be
4326 recorded - for instance if somebody sent an entire SpamAssassin report back
4327 as a large number of 550 error lines. This bug was coincidentally fixed by
4328 increasing the size of one of Exim's internal buffers (big_buffer) that
4329 happened as part of the Exiscan merge. However, to be on the safe side, I
4330 have made the code more robust (and fixed the comments that describe what
4333 33. Some experimental protocols are using DNS PTR records for new purposes. The
4334 keys for these records are domain names, not reversed IP addresses. The
4335 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4336 leaves it alone. Component reversal etc. now happens only for IP addresses.
4339 34. The host_aton() function is supposed to be passed a string that is known
4340 to be a valid IP address. However, in the case of IPv6 addresses, it was
4341 not checking this. This is a hostage to fortune. Exim now panics and dies
4342 if the condition is not met. A case was found where this could be provoked
4343 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4344 components; fortuitously, this particular loophole had already been fixed
4345 by change 4.50/55 or 4.44/33 above.
4347 If there are any other similar loopholes, the new check in host_aton()
4348 itself should stop them being exploited. The report I received stated that
4349 data on the command line could provoke the exploit when Exim was running as
4350 exim, but did not say which command line option was involved. All I could
4351 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4352 running as the user.
4355 35. There was a buffer overflow vulnerability in the SPA authentication code
4356 (which came originally from the Samba project). I have added a test to the
4357 spa_base64_to_bits() function which I hope fixes it.
4360 36. The daemon start-up calls getloadavg() while still root for those OS that
4361 need the first call to be done as root, but it missed one case: when
4362 deliver_queue_load_max is set with deliver_drop_privilege. This is
4363 necessary for the benefit of the queue runner, because there is no re-exec
4364 when deliver_drop_privilege is set.
4366 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4367 was in use, was not putting the data itself into the right store pool;
4368 consequently, it could be overwritten for a subsequent message in the same
4369 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4372 38. Sometimes the final signoff response after QUIT could fail to get
4373 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4374 < 0 before doing a fflush(). This bug looks as though it goes back to the
4375 introduction of TLS in release 3.20, but "sometimes" must have been rare
4376 because the tests only now provoked it.
4382 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4383 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4384 Now it returns an integer. A number of calls were still expecting a T/F
4385 return. Fortuitously, in all cases, the tests worked in OK situations,
4386 which is the norm. However, things would have gone wrong on any write
4387 failures on the smtp file descriptor. This function is used when sending
4388 messages over SMTP and also when doing verify callouts.
4390 2. When Exim is called to do synchronous delivery of a locally submitted
4391 message (the -odf or -odi options), it no longer closes stderr before doing
4394 3. Implemented the mua_wrapper option.
4396 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4398 5. Implemented the functions header_remove(), header_testname(),
4399 header_add_at_position(), and receive_remove_recipient(), and exported them
4402 6. If an ACL "warn" statement specified the addition of headers, Exim already
4403 inserted X-ACL-Warn: at the start if there was no header name. However, it
4404 was not making this test for the second and subsequent header lines if
4405 there were newlines in the string. This meant that an invalid header could
4406 be inserted if Exim was badly configured.
4408 7. Allow an ACL "warn" statement to add header lines at the start or after all
4409 the Received: headers, as well as at the end.
4411 8. Added the rcpt_4xx retry error code.
4413 9. Added postmaster_mailfrom=xxx to callout verification option.
4415 10. Added mailfrom=xxxx to the callout verification option, for verify=
4418 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4419 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4421 12. Inserted some casts to stop certain compilers warning when using pointer
4422 differences as field lengths or precisions in printf-type calls (mostly
4423 affecting debugging statements).
4425 13. Added optional readline() support for -be (dynamically loaded).
4427 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4428 same clock tick as a message's arrival, so that its received time was the
4429 same as the "first fail" time on the retry record, and that message
4430 remained on the queue past the ultimate address timeout, every queue runner
4431 would try a delivery (because it was past the ultimate address timeout) but
4432 after another failure, the ultimate address timeout, which should have then
4433 bounced the address, did not kick in. This was a "< instead of <=" error;
4434 in most cases the first failure would have been in the next clock tick
4435 after the received time, and all would be well.
4437 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4438 being recognized when the domain list was tested by the match_domain
4439 condition in an expansion string.
4441 16. Added the ${str2b64: operator.
4443 17. Exim was always calling setrlimit() to set a large limit for the number of
4444 processes, without checking whether the existing limit was already
4445 adequate. (It did check for the limit on file descriptors.) Furthermore,
4446 errors from getrlimit() and setrlimit() were being ignored. Now they are
4447 logged to the main and panic logs, but Exim does carry on, to try to do its
4448 job under whatever limits there are.
4450 18. Imported PCRE 5.0.
4452 19. Trivial typo in log message " temporarily refused connection" (the leading
4455 20. If the log selector return_path_on_delivery was set and an address was
4456 redirected to /dev/null, the delivery process crashed because it assumed
4457 that a return path would always be set for a "successful" delivery. In this
4458 case, the whole delivery is bypassed as an optimization, and therefore no
4461 21. Internal re-arrangement: the function for sending a challenge and reading
4462 a response while authentication was assuming a zero-terminated challenge
4463 string. It's now changed to take a pointer and a length, to allow for
4464 binary data in such strings.
4466 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4468 23. Exim was not respecting finduser_retries when seeking the login of the
4469 uid under which it was called; it was always trying 10 times. (The default
4470 setting of finduser_retries is zero.) Also, it was sleeping after the final
4471 failure, which is pointless.
4473 24. Implemented tls_on_connect_ports.
4475 25. Implemented acl_smtp_predata.
4477 26. If the domain in control=submission is set empty, Exim assumes that the
4478 authenticated id is a complete email address when it generates From: or
4479 Sender: header lines.
4481 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4482 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4483 chown and chgrp in /bin and hostname in /usr/bin.
4485 28. Exim was keeping the "process log" file open after each use, just as it
4486 does for the main log. This opens the possibility of it remaining open for
4487 long periods when the USR1 signal hits a daemon. Occasional processlog
4488 errors were reported, that could have been caused by this. Anyway, it seems
4489 much more sensible not to leave this file open at all, so that is what now
4492 29. The long-running daemon process does not normally write to the log once it
4493 has entered its main loop, and it closes the log before doing so. This is
4494 so that log files can straightforwardly be renamed and moved. However,
4495 there are a couple of unusual error situations where the daemon does write
4496 log entries, and I had neglected to close the log afterwards.
4498 30. The text of an SMTP error response that was received during a remote
4499 delivery was being truncated at 512 bytes. This is too short for some of
4500 the long messages that one sometimes sees. I've increased the limit to
4503 31. It is now possible to make retry rules that apply only when a message has a
4504 specific sender, in particular, an empty sender.
4506 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4507 it possible to be selective about when SMTP synchronization is enforced.
4509 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4511 32. Implemented hosts_connection_nolog.
4513 33. Added an ACL for QUIT.
4515 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4518 35. Added mailbox_size and mailbox_filecount to appendfile.
4520 36. Added control = no_multiline_responses to ACLs.
4522 37. There was a bug in the logic of the code that waits for the clock to tick
4523 in the case where the clock went backwards by a substantial amount such
4524 that the microsecond fraction of "now" was more than the microsecond
4525 fraction of "then" (but the whole seconds number was less).
4527 38. Added support for the libradius Radius client library this is found on
4528 FreeBSD (previously only the radiusclient library was supported).
4534 1. When certain lookups returned multiple values in the form name=value, the
4535 quoting of the values was not always being done properly. Specifically:
4536 (a) If the value started with a double quote, but contained no whitespace,
4538 (b) If the value contained whitespace other than a space character (i.e.
4539 tabs or newlines or carriage returns) it was not quoted.
4540 This fix has been applied to the mysql and pgsql lookups by writing a
4541 separate quoting function and calling it from the lookup code. The fix
4542 should probably also be applied to nisplus, ibase and oracle lookups, but
4543 since I cannot test any of those, I have not disturbed their existing code.
4545 2. A hit in the callout cache for a specific address caused a log line with no
4546 reason for rejecting RCPT. Now it says "Previous (cached) callout
4547 verification failure".
4549 3. There was an off-by-one bug in the queryprogram router. An over-long
4550 return line was truncated at 256 instead of 255 characters, thereby
4551 overflowing its buffer with the terminating zero. As well as fixing this, I
4552 have increased the buffer size to 1024 (and made a note to document this).
4554 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4555 when Exim is waiting for an SMTP response from a remote server, Exim
4556 restarts its select() call on the socket, thereby resetting its timeout.
4557 This is not a problem when such interrupts are rare. Somebody set up a cron
4558 job to run exiwhat every 2 minutes, which is less than the normal select()
4559 timeout (5 or 10 minutes). This meant that the select() timeout never
4560 kicked in because it was always reset. I have fixed this by comparing the
4561 time when an interrupt arrives with the time at the start of the first call
4562 to select(). If more time than the timeout has elapsed, the interrupt is
4563 treated as a timeout.
4565 5. Some internal re-factoring in preparation for the addition of Sieve
4566 extensions (by MH). In particular, the "personal" test is moved to a
4567 separate function, and given an option for scanning Cc: and Bcc: (which is
4568 not set for Exim filters).
4570 6. When Exim created an email address using the login of the caller as the
4571 local part (e.g. when creating a From: or Sender: header line), it was not
4572 quoting the local part when it contained special characters such as @.
4574 7. Installed new OpenBSD configuration files.
4576 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4577 try to make them clearer.
4579 9. Callout options, other than the timeout value, were being ignored when
4580 verifying sender addresses in header lines. For example, when using
4582 verify = header_sender/callout=no_cache
4584 the cache was (incorrectly) being used.
4586 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4587 only the Cygwin environment.
4589 11. When return_path_on_delivery was set as a log selector, if different remote
4590 addresses in the same message used different return paths and parallel
4591 remote delivery occurred, the wrong values would sometimes be logged.
4592 (Whenever a remote delivery process finished, the return path value from
4593 the most recently started remote delivery process was logged.)
4595 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4596 lines when AUTH and/or TLS are in use. This is the "received protocol"
4597 field. Exim used to use "asmtp" for authenticated SMTP, without any
4598 indication (in the protocol name) for TLS use. Now it follows the RFC and
4599 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4600 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4601 names appear in log lines as well as in Received: header lines.
4603 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4604 extensions, and comparison tests, and to fix some bugs.
4606 14. Changes to the "personal" filter test:
4608 (1) The test was buggy in that it was just doing the equivalent of
4609 "contains" tests on header lines. For example, if a user's address was
4610 anne@some.where, the "personal" test would incorrectly be true for
4612 To: susanne@some.where
4614 This test is now done by extracting each address from the header in turn,
4615 and checking the entire address. Other tests that are part of "personal"
4616 are now done using regular expressions (for example, to check local parts
4617 of addresses in From: header lines).
4619 (2) The list of non-personal local parts in From: addresses has been
4620 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4621 taken from the Sieve specification recommendations.
4623 (3) If the message contains any header line starting with "List-" it is
4624 treated as non-personal.
4626 (4) The test for "circular" in the Subject: header line has been removed
4627 because it now seems ill-conceived.
4629 15. Minor typos in src/EDITME comments corrected.
4631 16. Installed latest exipick from John Jetmore.
4633 17. If headers_add on a router specified a text string that was too long for
4634 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4635 of string_sprintf() is now avoided.
4637 18. $message_body_size was not set (it was always zero) when running the DATA
4638 ACL and the local_scan() function.
4640 19. For the "mail" command in an Exim filter, no default was being set for
4641 the once_repeat time, causing a random time value to be used if "once" was
4642 specified. (If the value happened to be <= 0, no repeat happened.) The
4643 default is now 0s, meaning "never repeat". The "vacation" command was OK
4644 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4645 (I found it when inspecting the code).
4647 20. There is now an overall timeout for performing a callout verification. It
4648 defaults to 4 times the callout timeout, which applies to individual SMTP
4649 commands during the callout. The overall timeout applies when there is more
4650 than one host that can be tried. The timeout is checked before trying the
4651 next host. This prevents very long delays if there are a large number of
4652 hosts and all are timing out (e.g. when the network connections are timing
4653 out). The value of the overall timeout can be changed by specifying an
4654 additional sub-option for "callout", called "maxwait". For example:
4656 verify = sender/callout=5s,maxwait=20s
4658 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4659 to the end before writing, but this should make it even safer).
4661 22. Exim was forgetting that it had advertised PIPELINING for the second and
4662 subsequent messages on an SMTP connection. It was also not resetting its
4663 memory on STARTTLS and an internal HELO.
4665 23. When Exim logs an SMTP synchronization error within a session, it now
4666 records whether PIPELINING has been advertised or not.
4668 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4669 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4670 rather than long int.
4672 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4674 26. Added the never_mail option to autoreply.
4680 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4681 crash if the getsockname() call failed; this can happen if a connection is
4682 closed very soon after it is established. The problem was simply in the
4683 order in which certain operations were done, causing Exim to try to write
4684 to the SMTP stream before it had set up the file descriptor. The bug has
4685 been fixed by making things happen in the correct order.
4691 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4692 before the connection was closed, thus losing the rejection response.
4694 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4695 some early Solaris releases, but causes trouble in current releases where
4696 socklen_t is defined.
4698 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4701 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4704 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4705 that releasing the top of it at the end releases what was used for sub-
4706 expansions (unless the block got too big). However, discard this block if
4707 the first thing is a variable or header, so that we can use its block when
4708 it is dynamic (useful for very large $message_headers, for example).
4710 6. Lookups now cache *every* query, not just the most recent. A new, separate
4711 store pool is used for this. It can be recovered when all lookup caches are
4712 flushed. Lookups now release memory at the end of their result strings.
4713 This has involved some general refactoring of the lookup sources.
4715 7. Some code has been added to the store_xxx() functions to reduce the amount
4716 of flapping under certain conditions.
4718 8. log_incoming_interface used to affect only the <= reception log lines. Now
4719 it causes the local interface and port to be added to several more SMTP log
4720 lines, for example "SMTP connection from", and rejection lines.
4722 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4724 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4726 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4727 could be overwritten at the end of the current message (or the start of a
4728 new message if it was set in a HELO ACL). The value is now preserved for
4729 the duration of the SMTP connection.
4731 12. If a transport had a headers_rewrite setting, and a matching header line
4732 contained an unqualified address, that address was qualified, even if it
4733 did not match any rewriting rules. The underlying bug was that the values
4734 of the flags that permit the existence of unqualified sender and recipient
4735 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4736 non-local messages, and by -bnq for local messages) were not being
4737 preserved with the message after it was received.
4739 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4740 "next input=" as part of the text comprising the host identity instead of
4741 the correct text. The code was using the same buffer for two different
4742 strings. However, depending on which order the printing function evaluated
4743 its arguments, the bug did not always show up. Under Linux, for example, my
4744 test suite worked just fine.
4746 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4747 doesn't work with some older versions of Perl. It has been changed to "my",
4748 which in any case is probably the better facility to use.
4750 15. A really picky compiler found some instances of statements for creating
4751 error messages that either had too many or two few arguments for the format
4754 16. The size of the buffer for calls to the DNS resolver has been increased
4755 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4756 for addresses that have a lot of PTR records. This alleviates a problem; it
4757 does not fully solve it.
4759 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4760 buffer now truncates the list and logs the incident, which is the same
4761 action as happens when Exim is looking up a host name and its aliases.
4762 Previously in this situation something unpredictable would happen;
4763 sometimes it was "internal error: store_reset failed".
4765 18. If a server dropped the connection unexpectedly when an Exim client was
4766 using GnuTLS and trying to read a response, the client delivery process
4767 crashed while trying to generate an error log message.
4769 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4770 string, for example:
4772 warn message = H1: something\nH2: something
4774 the text was added as a single header line from Exim's point of view
4775 though it ended up OK in the delivered message. However, searching for the
4776 second and subsequent header lines using $h_h2: did not work. This has been
4777 fixed. Similarly, if a system filter added multiple headers in this way,
4778 the routers could not see them.
4780 20. Expanded the error message when iplsearch is called with an invalid key to
4781 suggest using net-iplsearch in a host list.
4783 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4784 ACLs are no longer actually imposed (and a message to that effect is
4787 22. If a "gecos" field in a passwd entry contained escaped characters, in
4788 particular, if it contained a \" sequence, Exim got it wrong when building
4789 a From: or a Sender: header from that name. A second bug also caused
4790 incorrect handling when an unquoted " was present following a character
4791 that needed quoting.
4793 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4794 was not being matched caselessly.
4796 24. Arranged for all hyphens in the exim.8 source to be escaped with
4799 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4800 properly. Recipient callout cache records were still being keyed to include
4801 the sender, even when use_sender was set false. This led to far more
4802 callouts that were necessary. The sender is no longer included in the key
4803 when use_sender is false.
4805 26. Added "control = submission" modifier to ACLs.
4807 27. Added the ${base62d: operator to decode base 62 numbers.
4809 28. dnsdb lookups can now access SRV records.
4811 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4812 the configuration file.
4814 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4815 (-v) mode. This makes the output for a verbose queue run more intelligible.
4817 31. Added a use_postmaster feature to recipient callouts.
4819 32. Added the $body_zerocount variable, containing the number of binary zero
4820 bytes in the message body.
4822 33. The time of last modification of the "new" subdirectory is now used as the
4823 "mailbox time last read" when there is a quota error for a maildir
4826 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4828 35. Added +ignore_unknown as a special item in host lists.
4830 36. Code for decoding IPv6 addresses in host lists is now included, even if
4831 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4832 address was recognized as an IP address, but was then not correctly decoded
4833 into binary, causing unexpected and incorrect effects when compared with
4840 1. Very minor rewording of debugging text in manualroute to say "list of
4841 hosts" instead of "hostlist".
4843 2. If verify=header_syntax was set, and a header line with an unqualified
4844 address (no domain) and a large number of spaces between the end of the
4845 name and the colon was received, the reception process suffered a buffer
4846 overflow, and (when I tested it) crashed. This was caused by some obsolete
4847 code that should have been removed. The fix is to remove it!
4849 3. When running in the test harness, delay a bit after writing a bounce
4850 message to get a bit more predictability in the log output.
4852 4. Added a call to search_tidyup() just before forking a reception process. In
4853 theory, someone could use a lookup in the expansion of smtp_accept_max_
4854 per_host which, without the tidyup, could leave open a database connection.
4856 5. Added the variables $recipient_data and $sender_data which get set from a
4857 lookup success in an ACL "recipients" or "senders" condition, or a router
4858 "senders" option, similar to $domain_data and $local_part_data.
4860 6. Moved the writing of debug_print from before to after the "senders" test
4863 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4864 problems for message scanning, either using a data ACL, or using
4865 local_scan() because the Received: header was not generated till after they
4866 were called (in order to set the time as the time of reception completion).
4867 I have revised the way this works. The header is now generated after the
4868 body is received, but before the ACL or local_scan() are called. After they
4869 are run, the timestamp in the header is updated.
4875 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4876 before starting a queue runner without re-exec. This happened only when
4877 deliver_drop_privilege was set or when the Exim user was set to root. The
4878 effect of the bug was that timeouts during subsequent deliveries caused
4879 crashes instead of being properly handled. The handler is now left at its
4880 default (and expected) setting.
4882 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4883 message, again when deliver_drop_privilege is set or Exim is run as root.
4884 The bug described in (1) was not present in this case, but the tidying up
4885 of the other signals was missing. I have made the two cases consistent.
4887 3. The ignore_target_hosts setting on a manualroute router was being ignored
4888 for hosts that were looked up using the /MX notation.
4890 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4893 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4894 operated on the sender address. After changing the $sender_address to <>
4895 for the sender address verify, Exim was re-instated it as the original
4896 (before rewriting) address, but remembering that it had rewritten it, so it
4897 wasn't rewriting it again. This bug also had the effect of breaking the
4898 sender address verification caching when the sender address was rewritten.
4900 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4901 This has been changed so that if the ip literal address matches
4902 ignore_target_hosts, the router declines.
4904 7. Added expansion conditions match_domain, match_address, and match_local_
4905 part (NOT match_host).
4907 8. The placeholder for the Received: header didn't have a length field set.
4909 9. Added code to Exim itself and to exim_lock to test for a specific race
4910 condition that could lead to file corruption when using MBX delivery. The
4911 issue is with the lockfile that is created in /tmp. If this file is removed
4912 after a process has opened it but before that process has acquired a lock,
4913 there is the potential for a second process to recreate the file and also
4914 acquire a lock. This could lead to two Exim processes writing to the file
4915 at the same time. The added code performs the same test as UW imapd; it
4916 checks after acquiring the lock that its file descriptor still refers to
4917 the same named file.
4919 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4920 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4923 11. Added the smtp_active_hostname option. If used, this will typically be made
4924 to depend on the incoming interface address. Because $interface_address is
4925 not set up until the daemon has forked a reception process, error responses
4926 that can happen earlier (such as "too many connections") no longer contain
4929 12. If an expansion in a condition on a "warn" statement fails because a lookup
4930 defers, the "warn" statement is abandoned, and the next ACL statement is
4931 processed. Previously this caused the whole ACL to be aborted.
4933 13. Added the iplsearch lookup type.
4935 14. Added ident_timeout as a log selector.
4937 15. Added tls_certificate_verified as a log selector.
4939 16. Added a global option tls_require_ciphers (compare the smtp transport
4940 option of the same name). This controls incoming TLS connections.
4942 17. I finally figured out how to make tls_require_ciphers do a similar thing
4943 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4944 before starting the TLS session.
4946 18. Tabs are now shown as \t in -bP output.
4948 19. If the log selector return_path_on_delivery was set, Exim crashed when
4949 bouncing a message because it had too many Received: header lines.
4951 20. If two routers both had headers_remove settings, and the first one included
4952 a superfluous trailing colon, the final name in the first list and the
4953 first name in the second list were incorrectly joined into one item (with a
4954 colon in the middle).
4960 1. Added -C and -D options to the exinext utility, mainly to make it easier
4961 to include in the automated testing, but these could be helpful when
4962 multiple configurations are in use.
4964 2. The exinext utility was not formatting the output nicely when there was
4965 an alternate port involved in the retry record key, nor when there was a
4966 message id as well (for retries that were specific to a specific message
4967 and a specific host). It was also confused by IPv6 addresses, because of
4968 the additional colons they contain. I have fixed the IPv4 problem, and
4969 patched it up to do a reasonable job for IPv6.
4971 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4972 delivery, the log line now contains "pipelined" if PIPELINING was used.
4974 4. An SMTP transport process used to panic and die if the bind() call to set
4975 an explicit outgoing interface failed. This has been changed; it is now
4976 treated in the same way as a connect() failure.
4978 5. A reference to $sender_host_name in the part of a conditional expansion
4979 that was being skipped was still causing a DNS lookup. This no longer
4982 6. The def: expansion condition was not recognizing references to header lines
4983 that used bh_ and bheader_.
4985 7. Added the _cache feature to named lists.
4987 8. The code for checking quota_filecount in the appendfile transport was
4988 allowing one more file than it should have been.
4990 9. For compatibility with Sendmail, the command line option
4998 and sets the incoming protocol and host name (for trusted callers). The
4999 host name and its colon can be omitted when only the protocol is to be set.
5000 Note the Exim already has two private options, -pd and -ps, that refer to
5001 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5002 "s", but I don't think that's a major issue.
5004 10. A number of refactoring changes to the code, none of which should affect
5007 (a) The number of logging options was getting close to filling up the
5008 32-bit word that was used as a bit map. I have split them into two classes:
5009 those that are passed in the argument to log_write(), and those that are
5010 only ever tested independently outside of that function. These are now in
5011 separate 32-bit words, so there is plenty of room for expansion again.
5012 There is no change in the user interface or the logging behaviour.
5014 (b) When building, for example, log lines, the code previously used a
5015 macro that called string_cat() twice, in order to add two strings. This is
5016 not really sufficiently general. Furthermore, there was one instance where
5017 it was actually wrong because one of the argument was used twice, and in
5018 one call a function was used. (As it happened, calling the function twice
5019 did not affect the overall behaviour.) The macro has been replaced by a
5020 function that can join an arbitrary number of extra strings onto a growing
5023 (c) The code for expansion conditions now uses a table and a binary chop
5024 instead of a serial search (which was left over from when there were very
5025 few conditions). Also, it now recognizes conditions like "pam" even when
5026 the relevant support is not compiled in: a suitably worded error message is
5027 given if an attempt is made to use such a condition.
5029 11. Added ${time_interval:xxxxx}.
5031 12. A bug was causing one of the ddress fields not to be passed back correctly
5032 from remote delivery subprocesses. The field in question was not being
5033 subsequently used, so this caused to problems in practice.
5035 13. Added new log selectors queue_time and deliver_time.
5037 14. Might have fixed a bug in maildirsizefile handling that threw up
5038 "unexpected character" debug warnings, and recalculated the data
5039 unnecessarily. In any case, I expanded the warning message to give more
5042 15. Added the message "Restricted characters in address" to the statements in
5043 the default ACL that block characters like @ and % in local parts.
5045 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5046 Three changes have been made:
5048 (a) There was a serious bug; a negative response to MAIL caused the whole
5049 recipient domain to be cached as invalid, thereby blocking all messages
5050 to all local parts at the same domain, from all senders. This bug has
5051 been fixed. The domain is no longer cached after a negative response to
5052 MAIL if the sender used is not empty.
5054 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5057 (c) A new callout option, "use_sender" has been added for people who want
5058 the modified behaviour.
5064 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5067 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5068 indeed breaks things for older releases.
5070 3. Added additional logging to the case where there is a problem reading data
5071 from a filter that is running in a subprocess using a pipe, in order to
5072 try to track down a specific problem.
5074 4. Testing facility fudge: when running in the test harness and attempting
5075 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5076 getting "No route to host". Convert this to a timeout.
5078 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5081 6. Some OS don't have socklen_t but use size_t instead. This affects the
5082 fifth argument of getsockopt() amongst other things. This is now
5083 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5084 can be set for individual OS. I have set it for SunOS5, OSF1, and
5085 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5086 some earlier ones do not.
5088 7. Change 4.30/15 was not doing the test caselessly.
5090 8. The standard form for an IPv6 address literal was being rejected by address
5091 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5092 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5093 this, as well as the form without the "IPv6" on the front (but only when
5094 address literals are enabled, of course).
5096 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5098 10. Exim crashed if a message with an empty sender address specified by -f
5099 encountered a router with an errors_to setting. This could be provoked only
5100 by a command such as
5104 where an empty string was supplied; "<>" did not hit this bug.
5106 11. Installed PCRE release 4.5.
5108 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5109 remained set. It is now erased.
5111 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5112 times from message ids (which are base 36 rather than the normal 62).
5114 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5115 were being counted as actual protocol errors, and logged if the log
5116 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5117 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5118 and DATA following a set of rejected RCPTs do not count as protocol errors.
5119 In other words, Exim assumes they were pipelined, though this may not
5120 actually be the case. Of course, in all cases the client gets an
5121 appropriate error code.
5123 15. If a lookup fails in an ACL condition, a message about the failure may
5124 be available; it is used if testing the ACL cannot continue, because most
5125 such messages specify what the cause of the deferral is. However, some
5126 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5127 that caused an old message to be retained and used if a later statement
5128 caused a defer, replacing the real cause of the deferral.
5130 16. If an IP address had so many PTR records that the DNS lookup buffer
5131 was not large enough to hold them, Exim could crash while trying to process
5132 the truncated data. It now detects and logs this case.
5134 17. Further to 4.21/58, another change has been made: if (and only if) the
5135 first line of a message (the first header line) ends with CRLF, a bare LF
5136 in a subsequent header line has a space inserted after it, so as not to
5137 terminate the header.
5139 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5140 unnecessarily, used atoi() instead of strtol(), and didn't check the
5141 termination when getting file sizes from file names by regex.
5143 19. Completely re-implemented the support for maildirsize files, in the light
5144 of a number of problems with the previous contributed implementation
5145 (4.30/29). In particular:
5147 . If the quota is zero, the maildirsize file is maintained, but no quota is
5150 . If the maildir directory does not exist, it is created before any attempt
5151 to write a maildirsize file.
5153 . The quota value in the file is just a cache; if the quota is changed in
5154 the transport, the new value overrides.
5156 . A regular expression is available for excluding directories from the
5159 20. The autoreply transport checks the characters in options that define the
5160 message's headers; it allows continued headers, but it was checking with
5161 isspace() after an embedded newline instead of explicitly looking for a
5164 21. If all the "regular" hosts to which an address was routed had passed their
5165 expiry times, and had not reached their retry times, the address was
5166 bounced, even if fallback hosts were defined. Now Exim should go on to try
5169 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5170 equivalent code in the SMTP transport. Some hosts send humungous responses
5171 to HELO/EHLO, more than 1024 it seems.
5173 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5174 gives compiler warnings in some environments. I've now done it "properly",
5177 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5178 (because of problems with the built-in one) was declared to return uschar *
5179 instead of char *, causing compiler failure.
5181 25. Fixed a file descriptor leak when processing alias/forward files.
5183 26. Fixed a minor format string issue in dbfn.c.
5185 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5187 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5188 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5191 29. When a sender address is verified, it is cached, to save repeating the test
5192 when there is more than one recipient in a message. However, when the
5193 verification involves a callout, it is possible for different callout
5194 options to be set for different recipients. It is too complicated to keep
5195 track of this in the cache, so now Exim always runs a verification when a
5196 callout is required, relying on the callout cache for the optimization.
5197 The overhead is duplication of the address routing, but this should not be
5200 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5201 to be verified with callout=postmaster, and the main callout worked but the
5202 postmaster check failed, the verification correctly failed. However, if a
5203 subsequent RCPT command asked for sender verification *without* the
5204 postmaster check, incorrect caching caused this verification also to fail,
5207 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5208 it was not caching the DNS options (qualify_single, search_parents) that
5209 were used when the lookup failed. A subsequent lookup with different
5210 options therefore always gave the same answer, though there were cases
5211 where it should not have. (Example: a "domains = !$mx_any" option on a
5212 dnslookup router: the "domains" option is always processed without any
5213 widening, but the router might have qualify_single set.) Now Exim uses the
5214 cached value only when the same options are set.
5216 32. Added John Jetmore's "exipick" utility to the distribution.
5218 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5219 than a timeout (e.g. a certificate is required, and is not provided), an
5220 Exim server now closes the connection immediately. Previously it waited for
5221 the client to close - but if the client is SSL, it seems that they each
5222 wait for each other, leading to a delay before one of them times out.
5224 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5225 maintained 0.8.x compatibility because I don't think many are using it, and
5226 it is clearly obsolete.
5228 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5231 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5232 client certificate was expired. A simple patch fixes this, though I don't
5233 understand the full logic of why the verify callback is called multiple
5236 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5237 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5238 which causes problems with some clients (such as the Certicom SSL Plus
5239 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5240 disables the coutermeasure allowing Eudora to connect."
5242 38. Exim was not checking that a write() to a log file succeeded. This could
5243 lead to Bad Things if a log got too big, in particular if it hit a file
5244 size limit. Exim now panics and dies if it cannot write to a log file, just
5245 as it does if it cannot open a log file.
5247 39. Modified OS/Makefile-Linux so that it now contains
5249 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5251 The two -D definitions ensure that Exim is compiled with large file
5252 support, which makes it possible to handle log files that are bigger than
5255 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5256 instance) a domain was checked against a named list that involved a lookup,
5257 causing $domain_data to be set, then another domain was checked against the
5258 same list, then the first domain was re-checked, the value of $domain_data
5259 after the final check could be wrong. In particular, if the second check
5260 failed, it could be set empty. This bug probably also applied to
5263 41. The strip_trailing_dot option was not being applied to the address given
5264 with the -f command-line option.
5266 42. The code for reading a message's header from the spool was incrementing
5267 $received_count, but never initializing it. This meant that the value was
5268 incorrect (doubled) while delivering a message in the same process in which
5269 it was received. In the most common configuration of Exim, this never
5270 happens - a fresh exec is done - but it can happen when
5271 deliver_drop_privilege is set.
5273 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5274 it now includes up to 150 characters of the unexpected data in the log
5277 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5278 and building data strings. The size of both of these buffers was 10 000
5279 bytes - far larger than anybody would *ever* want, thought I. Needless to
5280 say, somebody hit the limit. I have increased the maximum line length to
5281 20 000 and the maximum data length of concatenated lines to 100 000. I have
5282 also fixed two bugs, because there was no checking on these buffers. Tsk,
5283 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5284 buffer is too small.
5286 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5287 lsearch lookups. Now it does.
5289 46. When parsing a route_list item in a manualroute router, a fixed-length
5290 buffer was used for the list of hosts. I made this 1024 bytes long,
5291 thinking that nobody would ever have a list of hosts that long. Wrong.
5292 Somebody had a whole pile of complicated expansion conditions, and the
5293 string was silently truncated, leading to an expansion error. It turns out
5294 that it is easier to change to an unlimited length (owing to other changes
5295 that have happened since this code was originally written) than to build
5296 structure for giving a limitation error. The length of the item that
5297 expands into the list of hosts is now unlimited.
5299 47. The lsearch lookup could not handle data where the length of text line was
5300 more than 4095 characters. Such lines were truncated, leading to shortened
5301 data being returned. It should now handle lines of any length.
5303 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5304 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5307 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5308 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5309 no longer happens in comment lines. A list of replacements is now placed
5310 at the head of all of the source files, except those whose only change is
5311 to replace PERL_COMMAND in the very first #! line.
5313 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5314 messages on the queue, with a bottom-up merge sort, using code contributed
5315 by Michael Haardt. This should make operations like -bp somewhat faster on
5316 large queues. It won't affect queue runners, except when queue_run_in_order
5319 51. Installed eximstats 1.31 in the distribution.
5321 52. Added support for SRV lookups to the dnslookup router.
5323 53. If an ACL referred to $message_body or $message_body_end, the value was not
5324 reset for any messages that followed in the same SMTP session.
5326 54. The store-handling optimization for building very long strings was not
5327 differentiating between the different store pools. I don't think this
5328 actually made any difference in practice, but I've tidied it.
5330 55. While running the routers to verify a sender address, $sender_address
5331 was still set to the sender address. This is wrong, because when routing to
5332 send a bounce to the sender, it would be empty. Therefore, I have changed
5333 it so that, while verifying a sender address, $sender_address is set to <>.
5334 (There is no change to what happens when verifying a recipient address.)
5336 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5337 target A or AAAA records (if not already returned) without resetting the
5338 qualify_single or search_parents options of the DNS resolver. These are
5339 inappropriate in this case because the targets of MX and SRV records must
5340 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5341 target that, when qualified, matched something in the local domain. These
5342 two options are now turned off when doing these lookups.
5344 57. It seems that at least some releases of Reiserfs (which does not have the
5345 concept of a fixed number of inodes) returns zero and not -1 for the
5346 number of available inodes. This interacted badly with check_spool_inodes,
5347 which assumed that -1 was the "no such thing" setting. What I have done is
5348 to check that the total number of inodes is greater than zero before doing
5349 the test of how many are available.
5351 58. When a "warn" ACL statement has a log_message modifier, the message is
5352 remembered, and not repeated. This is to avoid a lot of repetition when a
5353 message has many recipients that cause the same warning to be written.
5354 Howewer, Exim was preserving the list of already written lines for an
5355 entire SMTP session, which doesn't seem right. The memory is now reset if a
5356 new message is started.
5358 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5359 debugging output unless log_rewrite was also set.
5361 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5362 of (int)(handle) when we know that handle contains (void *)(-1).
5364 61. The Exim daemon panic-logs an error return when it closes the incoming
5365 connection. However "connection reset by peer" seems to be common, and
5366 isn't really an error worthy of noting specially, so that particular error
5369 62. When Exim is trying to find all the local interfaces, it used to panic and
5370 die if the ioctl to get the interface flags failed. However, it seems that
5371 on at least one OS (Solaris 9) it is possible to have an interface that is
5372 included in the list of interfaces, but for which you get a failure error
5373 for this call. This happens when the interface is not "plumbed" into a
5374 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5375 failure of the "get flags" call assumes that the interface is down.
5377 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5378 makes life easier for people who are doing arithmetic on fields extracted
5379 from dates, where you often get leading zeros that should not be
5380 interpreted as octal.
5382 64. Added qualify_domain to the redirect router, to override the global
5385 65. If a pathologically long header line contained very many addresses (the
5386 report of this problem mentioned 10 000) and each of them was rewritten,
5387 Exim could use up a very large amount of memory. (It kept on making new
5388 copies of the header line as it rewrote, and never released the old ones.)
5389 At the expense of a bit more processing, the header rewriting function has
5390 been changed so that it no longer eats memory in this way.
5392 66. The generation of the Received: header has been moved from the time that a
5393 message starts to be received, to the time that it finishes. The timestamp
5394 in the Received: header should now be very close to that of the <= log
5395 line. There are two side-effects of this change:
5397 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5398 logged header lines no longer include the local Received: line, because
5399 it has not yet been created. The same applies to a copy of the message
5400 that is returned to a non-SMTP sender when a message is rejected.
5402 (b) When a filter file is tested using -bf, no additional Received: header
5403 is added to the test message. After some thought, I decided that this
5406 This change does not affect the value of $received_for. It is still set
5407 after address rewriting, but before local_scan() is called.
5409 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5411 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5412 gave an unhelpful panic error message, and a defer error. I have managed to
5413 change this behaviour so that it now rejects any supplied certificate,
5414 which seems right, as the list of acceptable certificates is empty.
5416 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5417 gave an unhelpful defer error. I have not managed to make this reject any
5418 supplied certificates, but the error message it gives is "no certificate
5419 supplied", which is not helpful.
5421 70. exigrep's output now also includes lines that are not associated with any
5422 message, but which match the given pattern. Implemented by a patch from
5423 Martin Sluka, which also tidied up the Perl a bit.
5425 71. Recipient callout verification, like sender verification, was using <> in
5426 the MAIL FROM command. This isn't really the right thing, since the actual
5427 sender may affect whether the remote host accepts the recipient or not. I
5428 have changed it to use the actual sender in the callout; this means that
5429 the cache record is now keyed on a recipient/sender pair, not just the
5430 recipient address. There doesn't seem to be a real danger of callout loops,
5431 since a callout by the remote host to check the sender would use <>.
5432 [SEE ABOVE: changed after hitting problems.]
5434 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5435 temporary errors. However, in the case of such a code being given after
5436 the end of a data transmission (i.e. after ".") Exim was failing to write
5437 a retry record for the message. (Yes, there was some broken host that was
5438 actually sending 8xx at this point.)
5440 73. An unknown lookup type in a host list could cause Exim to panic-die when
5441 the list was checked. (An example that provoked this was putting <; in the
5442 middle of a list instead of at the start.) If this happened during a DATA
5443 ACL check, a -D file could be left lying around. This kind of configuration
5444 error no longer causes Exim to die; instead it causes a defer errror. The
5445 incident is still logged to the main and panic logs.
5447 74. Buglet left over from Exim 3 conversion. The message "too many messages
5448 in one connection" was written to the rejectlog but not the mainlog, except
5449 when address rewriting (yes!) was being logged.
5451 75. Added write_rejectlog option.
5453 76. When a system filter was run not as root (that is, when system_filter_user
5454 was set), the values of the $n variables were not being returned to the
5455 main process; thus, they were not subsequently available in the $sn
5458 77. Added +return_path_on_delivery log selector.
5460 78. A connection timeout was being treated differently from recipients deferred
5461 when testing hosts_max_try with a message that was older than the host's
5462 retry timeout. (The host should not be counted, thus allowing all hosts to
5463 be tried at least once before bouncing.) This may have been the cause of an
5464 occasionally reported bug whereby a message would remain on the queue
5465 longer than the retry timeout, but would be bounced if a delivery was
5466 forced. I say "may" because I never totally pinned down the problem;
5467 setting up timeout/retry tests is difficult. See also the next item.
5469 79. The ultimate address timeout was not being applied to errors that involved
5470 a combination of host plus message (for example, a timeout on a MAIL
5471 command). When an address resolved to a number of possible hosts, and they
5472 were not all tried for each delivery (e.g. because of hosts_max_try), a
5473 message could remain on the queue longer than the retry timeout.
5475 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5478 81. Fixed an obscure SMTP outgoing bug which required at least the following
5479 conditions: (a) there was another message waiting for the same server;
5480 (b) the server returned 5xx to all RCPT commands in the first message so
5481 that the message was not completed; (c) the server dropped the connection
5482 or gave a negative response to the RSET that Exim sends to abort the
5483 transaction. The observed case was a dropped connection after DATA that had
5484 been sent in pipelining mode. That is, the server had advertised PIPELINING
5485 but was not implementing it correctly. The effect of the bug was incorrect
5486 behaviour, such as trying another host, and this could lead to a crash.
5492 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5493 and daemon.c were passed as pointers to ints; they should have been
5494 pointers to socklen_t variables (which are typically unsigned ints).
5496 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5499 3. Fixed a really odd bug that affected only the testing scheme; patching a
5500 certain fixed string in the binary changed the value of another string that
5501 happened to be identical to the end of the original first string.
5503 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5504 name", it returns that address as the IP address. On some operating
5505 systems (e.g. Solaris), it also passes back the IP address string as the
5506 "host name". However, on others (e.g. Linux), it passes back an empty
5507 string. Exim wasn't checking for this, and was changing the host name to an
5508 empty string, assuming it had been canonicized.
5510 5. Although rare, it is permitted to have more than one PTR record for a given
5511 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5512 all the names associated with an address, because they do in Solaris.
5513 However, it seems that they do not in Linux for data that comes from the
5514 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5515 I found this out when I moved to a new Linux workstation and tried to run
5516 the Exim test suite.
5518 To get round this problem I have changed the code so that it now does its
5519 own call to the DNS to look up PTR records when searching for a host name.
5520 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5521 addresses that are only in /etc/hosts are still found.
5523 This behaviour is, however, controlled by an option called host_lookup_
5524 order, which defaults to "bydns:byaddr". If people want to use the other
5525 order, or indeed, just use one or the other means of lookup, they can
5526 specify it in this variable.
5528 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5529 some operating systems, this comes back from gethostbyaddr() as an empty
5530 string, and this is what Exim used to test for. However, it seems that in
5531 other systems, "." is yielded. Exim now tests for this case too.
5533 7. The values of check_spool_space and check_log_space are now held internally
5534 as a number of kilobytes instead of an absolute number of bytes. If a
5535 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5536 kilobyte. This means that much larger values can be stored.
5538 8. Exim monitor: an attempt to get the action menu when not actually pointing
5539 at a message produces an empty menu entitled "No message selected". This
5540 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5541 no entries in it ("Shell widget menu has zero width and/or height"). So I
5542 have added a single, blank menu entry in this case.
5544 9. Added ${quote_local_part.
5546 10. MIME decoding is now applied to the contents of Subject: header lines when
5549 11. Now that a reference to $sender_host_address automatically causes a reverse
5550 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5551 host lookup before query-style lookups in lists that might use this
5552 variable. This has therefore been abolished, and the "net-" prefix is no
5553 longer necessary for query-style lookups.
5555 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5556 have been a typo for LFLAGS, so it has been changed.
5558 13. The install script calls Exim with "-C /dev/null" in order to find the
5559 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5560 to be output. Howeve, since Exim outputs its version number before the
5561 error, it didn't break the script. It just looked ugly. I fixed this by
5562 always allowing "-C /dev/null" if the caller is root.
5564 14. Ignore overlarge ACL variable number when reading spool file - insurance
5565 against a later release with more variables having written the file.
5567 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5568 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5569 this, as well as the form without the "IPv6" on the front.
5571 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5572 OS/Makefile-Darwin file.
5574 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5575 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5577 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5578 variable. However, some OS use ints for inodes. I've added cast to long int
5579 to get rid of the compiler warning.
5581 19. I had forgotten to lock out "/../" in configuration file names when
5582 ALT_CONFIG_PREFIX was set.
5584 20. Routers used for verification do not need to specify transports. However,
5585 if such a router generated a host list, and callout was configured, Exim
5586 crashed, because it could not find a port number from the (non-existent)
5587 transport. It now assumes port 25 in this circumstance.
5589 21. Added the -t option to exigrep.
5591 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5592 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5593 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5594 them is set, LOOKUP_LSEARCH is forced.
5596 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5598 24. Added sender and host information to the "rejected by local_scan()" log
5599 line; previously there was no indication of these.
5601 25. Added .include_if_exists.
5603 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5604 Linux. It turns out that not all file systems support this. Apparently some
5605 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5606 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5607 not supported on the file descriptor, is now ignored when Exim is trying to
5608 sync a directory. This applies only to Linux.
5610 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5612 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5613 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5616 29. Added support for maildirsize files from supplied patch (modified a bit).
5618 30. The use of :fail: followed by an empty string could lead Exim to respond to
5619 sender verification failures with (e.g.):
5621 550 Verification failed for <xxx>
5622 550 Sender verify failed
5624 where the first response line was missing the '-' that indicates it is not
5625 the final line of the response.
5627 31. The loop for finding the name of the user that called Exim had a hardwired
5628 limit of 10; it now uses the value of finduser_retries, which is used for
5629 all other user lookups.
5631 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5634 33. Exim was neglecting to zero errno before one call of strtol() when
5635 expanding a string and expecting an integer value. On some systems this
5636 resulted in spurious "integer overflow" errors. Also, it was casting the
5637 result into an int without checking.
5639 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5640 did not work. The code looks as if it has *never* worked, though it appears
5641 to have been documented since at least releast 1.62. I have made it work.
5643 35. The "timeout_DNS" error in retry rules, also documented since at least
5644 1.62, also never worked. As it isn't clear exactly what this means, and
5645 clearly it isn't a major issue, I have abolished the feature by treating it
5646 as "timeout", and writing a warning to the main and panic logs.
5648 36. The display of retry rules for -brt wasn't always showing the error code
5651 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5652 timeout_connect_A, timeout_connect_MX.
5654 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5655 to the empty sender.
5657 39. The daemon was not analysing the content of -oX till after it had closed
5658 stderr and disconnected from the controlling terminal. This meant that any
5659 syntax errors were only noted on the panic log, and the return code from
5660 the command was 0. By re-arranging the code a little, I've made the
5661 decoding happen first, so such errors now appear on stderr, and the return
5662 code is 1. However, the actual setting up of the sockets still happens in
5663 the disconnected process, so errors there are still only recorded on the
5666 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5667 connections (as happens on some IP stacks) was logged at start up time as
5668 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5669 it from "IPv6 and IPv4", which means that two separate sockets are being
5672 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5673 says whether AF_INET or AF_INET6 was passed as an argument.
5675 42. Exiwhat output was messed up when time zones were included in log
5678 43. Exiwhat now gives more information about the daemon's listening ports,
5679 and whether -tls-on-connect was used.
5681 44. The "port" option of the smtp transport is now expanded.
5683 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5684 silently ignored. Now an error message is written to the main and panic
5687 46. There's a new ACL modifier called "logwrite" which writes to a log file
5688 as soon as it is encountered.
5690 47. Added $local_user_uid and $local_user_gid at routing time.
5692 48. Exim crashed when trying to verify a sender address that was being
5695 49. Exim was recognizing only a space character after ".include". It now also
5696 recognizes a tab character.
5698 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5699 extracting the relevant information from the specification. The man page no
5700 longer contains scrambled data for the -d option, and I've added a section
5701 at the front about calling Exim under different names.
5703 51. Added "extra_headers" argument to the "mail" command in filter files.
5705 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5708 53. Installed eximstats 1.29.
5710 54. Added transport_filter_timeout as a generic transport option.
5712 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5713 Cc: header lines. This was required by RFC 822, but it not required by RFC
5716 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5717 incoming messages that did not have them. Now it does so only if the
5718 message originates locally, that is, if there is no associated remote host
5719 address. When Resent- header lines are present, this applies to the Resent-
5720 lines rather than the non-Resent- lines.
5722 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5723 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5725 58. Messages for configuration errors now include the name of the main
5726 configuration file - useful now that there may be more than one file in a
5727 list (.included file names were always shown).
5729 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5730 for those rare installations that do not start the daemon as root or run it
5731 setuid root. I've cut out the call to initgroups() if the daemon is not
5734 60. The Exim user and group can now be bound into the binary as text strings
5735 that are looked up at the start of Exim's processing.
5737 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5739 62. Added $mailstore_basename variable.
5741 63. Installed patch to sieve.c from Michael Haardt.
5743 64. When Exim failed to open the panic log after failing to open the main log,
5744 the original message it was trying to log was written to stderr and debug
5745 output, but if they were not available (the usual case in production), it
5746 was lost. Now it is written to syslog before the two lines that record the
5747 failures to open the logs.
5749 65. Users' Exim filters run in subprocesses under the user's uid. It is
5750 possible for a "deliver" command or an alias in a "personal" command to
5751 provoke an address rewrite. If logging of address rewriting is configured,
5752 this fails because the process is not running as root or exim. There may be
5753 a better way of dealing with this, but for the moment (because 4.30 needs
5754 to be released), I have disabled address rewrite logging when running a
5755 filter in a non-root, non-exim process.
5761 1. The buildconfig auxiliary program wasn't quoting the value set for
5762 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5763 not defined. This bug was masked in 4.22 by the effect that was fixed in
5766 2. Some messages that were rejected after a message id was allocated were
5767 shown as "incomplete" by exigrep. It no longer does this for messages that
5768 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5770 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5771 have allow_domain_literals set, the ID did not get logged in the <= line.
5772 Domain literals are now always recognized in Message-ID: header lines.
5774 4. The first argument for a ${extract expansion item is the key name or field
5775 number. Leading and trailing spaces in this item were not being ignored,
5776 causing some misleading effects.
5778 5. When deliver_drop_privilege was set, single queue runner processes started
5779 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5780 same command in the process it spins off) were not dropping privilege.
5782 6. When the daemon running as "exim" started a queue runner, it always
5783 re-executed Exim in the spun-off process. This is a waste of effort when
5784 deliver_drop_privilege is set. The new process now just calls the
5785 queue-runner function directly.
5791 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5794 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5795 ignored. Though the use of -oP was forcing the writing of a pid file, it
5796 was always written to the default place.
5798 3. If the message "no IP address found for host xxxx" is generated during
5799 incoming verification, it is now followed by identification of the incoming
5800 connection (so you can more easily find what provoked it).
5802 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5804 5. Added some features to "harden" Exim a bit more against certain attacks:
5806 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5807 be put in Local/Makefile. This is like the never_users runtime option,
5808 but it cannot be overridden. The default setting is "root".
5810 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5811 prefix string with which any file named in a -C command line option
5814 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5815 is retained for -C and -D only if the caller of Exim is root. Without
5816 it, the exim user may also use -C and -D and retain privilege.
5818 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5819 command line option is disabled.
5821 6. Macro names set by the -D option must start with an upper case letter, just
5822 like macro names defined in the configuration file.
5824 7. Added "dereference=" facility to LDAP.
5826 8. Two instances of the typo "uknown" in the source files are fixed.
5828 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5829 the Configure-Makefile script screwed up while processing it.
5831 10. Incorporated PCRE 4.4.
5833 11. The SMTP synchronization check was not operating right at the start of an
5834 SMTP session. For example, it could not catch a HELO sent before the client
5835 waited for the greeting. There is now a check for outstanding input at the
5836 point when the greeting is written. Because of the duplex, asynchronous
5837 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5838 way, but not yet received, when the check is performed.
5840 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5841 on TCP/IP sockets, because this apparently causes some broken clients to
5844 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5845 unchanged) from the Cygwin maintainer.
5847 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5848 was supported instead of testing for mbx. Effectively a typo.
5850 15. The spa authenticator server code was not checking that the input it
5851 received was valid base64.
5853 16. The debug output line for the "set" modifier in ACLs was not showing the
5854 name of the variable that was being set.
5856 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5858 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5859 lookup on its own. Something else was needed to trigger the lookup. For
5860 example, a match in host_lookup or the need for a host name in a host list.
5861 Now, if $sender_host_name is referenced and the host name has not yet been
5862 looked up, a lookup is performed. If the lookup fails, the variable remains
5863 empty, and $host_lookup_failed is set to "1".
5865 19. Added "eqi" as a case-independent comparison operator.
5867 20. The saslauthd authentication condition could segfault if neither service
5868 nor realm was specified.
5870 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5871 error message that was logged was misleading, and incoming SMTP
5872 connections were dropped. The message is now more accurate, and temporary
5873 errors are given to SMTP connections.
5875 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5876 (except RSET) with a 421 error, until QUIT is received. However, it was
5877 failing to send a response to QUIT.
5879 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5880 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5881 ACL is now run after the helo_try_verify_hosts code.
5883 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5884 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5885 case-independent, so other case variants are also recognized). Apparently
5886 some systems use these upper case variants.
5888 25. If more than two messages were waiting for the same host, and a transport
5889 filter was specified for the transport, Exim sent two messages over the
5890 same TCP/IP connection, and then failed with "socket operation on non-
5891 socket" when it tried to send the third.
5893 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5895 27. The extern definition of crypt16() in expand.c was not being excluded when
5896 the OS had its own crypt16() function.
5898 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5899 as a preferred synonym for return_size_limit, both as an option and as an
5902 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5904 30. Changed the default configuration ACL to relax the local part checking rule
5905 for addresses that are not in any local domains. For these addresses,
5906 slashes and pipe symbols are allowed within local parts, but the sequence
5907 /../ is explicitly forbidden.
5909 31. SPA server authentication was not clearing the challenge buffer before
5912 32. log_message in a "warn" ACL statement was writing to the reject log as
5913 well as to the main log, which contradicts the documentation and doesn't
5914 seem right (because no rejection is happening). So I have stopped it.
5916 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5917 However, I am unable to do any testing of this.
5919 34. Fixed an infelicity in the appendfile transport. When checking directories
5920 for a mailbox, to see if any needed to be created, it was accidentally
5921 using path names with one or more superfluous leading slashes; tracing
5922 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5924 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5925 contents are added to the log line that is written for every discarded
5926 recipient. (Previously a log_message setting was ignored.)
5928 36. The ${quote: operator now quotes the string if it is empty.
5930 37. The install script runs exim in order to find its version number. If for
5931 some reason other than non-existence or emptiness, which it checks, it
5932 could not run './exim', it was installing it with an empty version number,
5933 i.e. as "exim-". This error state is now caught, and the installation is
5936 38. An argument was missing from the function that creates an error message
5937 when Exim fails to connect to the socket for saslauthd authentication.
5938 This could cause Exim to crash, or give a corrupted message.
5940 39. Added isip, isip4, and isip6 to ${if conditions.
5942 40. The ACL variables $acl_xx are now saved with the message, and can be
5943 accessed later in routers, transports, and filters.
5945 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5946 strings in the file are not string-expanded.
5948 42. If a MAIL command specified a SIZE value that was too large to fit into an
5949 int variable, the check against message_size_limit failed. Such values are
5950 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5951 day this will have to be increased, but I don't think I want to be around
5952 when emails are that large.
5959 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5960 iconv() is not standard in FreeBSD.
5962 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5963 IPv6 enabled. The observed symptom was a segmentation fault on return from
5964 the function os_common_find_running_interfaces() in src/os.c.
5966 3. In the check_special_case() function in daemon.c I had used "errno" as an
5967 argument name, which causes warnings on some systems. This was basically a
5968 typo, since it was named "eno" in the comments!
5970 4. The code that waits for the clock to tick (at a resolution of some fraction
5971 of a second) so as to ensure message-id uniqueness was always waiting for
5972 at least one whole tick, when it could have waited for less. [This is
5973 almost certainly not relevant at current processor speeds, where it is
5974 unlikely to ever wait at all. But we try to future-proof.]
5976 5. The function that sleeps for a time interval that includes fractions of a
5977 second contained a race. It did not block SIGALRM between setting the
5978 timer, and suspending (a couple of lines later). If the interval was short
5979 and the sigsuspend() was delayed until after it had expired, the suspension
5980 never ended. On busy systems this could lead to processes getting stuck for
5983 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5984 process, before it forks any delivery processes. The open lookup caching
5985 mechanism meant that the open file or database connection was passed into
5986 the delivery process. The problem was that delivery processes always tidy
5987 up cached lookup data. This could cause a problem for the next delivery
5988 process started by the queue runner, because the external queue runner
5989 process does not know about the closure. So the next delivery process
5990 still has data in the lookup cache. In the case of a file lookup, there was
5991 no problem because closing a file descriptor in a subprocess doesn't affect
5992 the parent. However, if the lookup was caching a connection to a database,
5993 the connection was closed, and the second delivery process was likely to
5994 see errors such as "PGSQL: query failed: server closed the connection
5995 unexpectedly". The problem has been fixed by closing all cached lookups
5996 in a queue runner before running a delivery process.
5998 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5999 seem to have the "const" qualifier which it has on other OS. I've
6002 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6003 *of the same type* (client or server) with the same public name that an
6004 error should be diagnosed.
6006 9. When Exim looked up a host name for an IP address, but failed to find the
6007 original IP address when looking up the host name (a safety check), it
6008 output the message "<ip address> does not match any IP for NULL", which was
6009 confusing, to say the least. The bug was that the host name should have
6010 appeared instead of "NULL".
6012 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6013 user that is built into the binary, and the -C or -D options is used, root
6014 privilege is dropped before the configuration file is read. In addition,
6015 logging is switched to stderr instead of the normal log files. If the
6016 configuration then re-defines the Exim user, the unprivileged environment
6017 is probably not what is expected, so Exim logs a panic warning message (but
6020 However, if deliver_drop_privilege is set, the unprivileged state may well
6021 be exactly what is intended, so the warning has been cut out in that case,
6022 and Exim is allowed to try to write to its normal log files.
6028 1. smtp_return_error_details was not giving details for temporary sender
6029 or receiver verification errors.
6031 2. Diagnose a configuration error if two authenticators have the same public
6034 3. Exim used not to create the message log file for a message until the first
6035 delivery attempt. This could be confusing when incoming messages were held
6036 for policy or load reasons. The message log file is now created at the time
6037 the message is received, and an initial "Received" line is written to it.
6039 4. The automatically generated man page for command line options had a minor
6040 bug that caused no ill effects; however, a more serious problem was that
6041 the procedure for building the man page automatically didn't always
6042 operate. Consequently, release 4.20 contains an out-of-date version. This
6043 shouldn't happen again.
6045 5. When building Exim with embedded Perl support, the script that builds the
6046 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6047 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6049 6. The freeze_tell option was not being used for messages that were frozen on
6050 arrival, either by an ACL or by local_scan().
6052 7. Added the smtp_incomplete_transaction log selector.
6054 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6055 was accepting AUTH without a new EHLO.
6057 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6058 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6061 10. Logging of TCP/IP connections (when configured) now happens in the main
6062 daemon process instead of the child process, so that the TCP/IP connection
6063 count is more accurate (but it can never be perfect).
6065 11. The use of "drop" in a nested ACL was not being handled correctly in the
6066 outer ACL. Now, if condition failure induced by the nested "drop" causes
6067 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6068 or "require"), the connection is dropped.
6070 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6071 that yield "discard" can now be used with an "accept" or a "discard" verb,
6072 but an error is generated for any others (because I can't see a useful way
6073 to define what should happen).
6075 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6076 are now processed in the same way as lines in the Exim configuration file.
6077 In particular, continuation lines are supported.
6079 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6081 15. Added -ti meaning -t -i.
6083 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6084 domains, and warn by logging if others are found.
6086 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6087 returned from GIFCONF when Exim is trying to find the list of interfaces on
6088 a host. The code in os.c has been modified to copy each ifreq to an aligned
6089 structure in all cases.
6091 Also, in some cases, the returned ifreq's were being copied to a 'struct
6092 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6093 means the last couple of bytes of an IPv6 address could be chopped if the
6094 ifreq contained only a normal sockaddr (14 bytes storage).
6096 18. Named domain lists were not supported in the hosts_treat_as_local option.
6097 An entry such as +xxxx was not recognized, and was treated as a literal
6100 19. Ensure that header lines added by a DATA ACL are included in the reject log
6101 if the ACL subsequently rejects the message.
6103 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6104 MD5 (which is deprecated).
6106 21. When testing a filter file using -bf, Exim was writing a message when it
6107 took the sender from a "From " line in the message, but it was not doing so
6108 when it took $return_path from a Return-Path: header line. It now does.
6110 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6111 with a valid header line field name (a series of printing characters
6112 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6114 23. Changed "disc" in the source to "disk" to conform to the documentation and
6115 the book and for uniformity.
6117 24. Ignore Sendmail's -Ooption=value command line item.
6119 25. When execve() failed while trying to run a command in a pipe transport,
6120 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6121 could be confused with a return value of 69 from the command itself. This
6122 has been changed to 127, the value the shell returns if it is asked to run
6123 a non-existent command. The wording for the related log line suggests a
6124 non-existent command as the problem.
6126 26. If received_header_text expands to an empty string, do not add a Received:
6127 header line to the message. (Well, it adds a token one on the spool, but
6128 marks it "old" so that it doesn't get used or transmitted.)
6130 27. Installed eximstats 1.28 (addition of -nt option).
6132 28. There was no check for failure on the call to getsockname() in the daemon
6133 code. This can fail if there is a shortage of resources on the system, with
6134 ENOMEM, for example. A temporary error is now given on failure.
6136 29. Contrary to the C standard, it seems that in some environments, the
6137 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6138 program. Exim now does this explicitly; it affects the formatting of
6139 timestamps using strftime().
6141 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6142 complaints. I've now initialized all the variables, to avoid this.
6144 32. Header lines added by a system filter were not being "seen" during
6145 transport-time rewrites.
6147 33. The info_callback() function passed to OpenSSL is set up with type void
6148 (*)(SSL *, int, int), as described somewhere. However, when calling the
6149 function (actually a macro) that sets it up, the type void(*)() is
6150 expected. I've put in a cast to prevent warnings from picky compilers.
6152 34. If a DNS black list lookup found a CNAME record, but there were no A
6153 records associated with the domain it pointed at, Exim crashed.
6155 35. If a DNS black list lookup returned more than one A record, Exim ignored
6156 all but the first. It now scans all returned addresses if a particular IP
6157 value is being sought. In this situation, the contents of the
6158 $dnslist_value variable are a list of all the addresses, separated by a
6161 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6162 to accept a host name and all its aliases if the forward lookup for any of
6163 them yielded the IP address of the incoming connection. Now it accepts only
6164 those names whose forward lookup yields the correct IP address. Any other
6165 names are discarded. This closes a loophole whereby a rogue DNS
6166 administrator could create reverse DNS records to break through a
6167 wildcarded host restriction in an ACL.
6169 37. If a user filter or a system filter that ran in a subprocess used any of
6170 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6171 the wrong values were passed to the pipe command ($thisaddress had the
6172 value of $0, $0 had the value of $1, etc). This bug was introduced by
6173 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6175 38. Improved the line breaking for long SMTP error messages from ACLs.
6176 Previously, if there was no break point between 40 and 75 characters, Exim
6177 left the rest of the message alone. Two changes have been made: (a) I've
6178 reduced the minimum length to 35 characters; (b) if it can't find a break
6179 point between 35 and 75 characters, it looks ahead and uses the first one
6180 that it finds. This may give the occasional overlong line, but at least the
6181 remaining text gets split now.
6183 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6184 file descriptors might be low, and that setting 1000 would always raise it.
6185 It turns out that in some environments, the limit is already over 1000 and
6186 that lowering it causes trouble. So now Exim takes care not to decrease it.
6188 40. When delivering a message, the value of $return_path is set to $sender_
6189 address at the start of routing (routers may change the value). By an
6190 oversight, this default was not being set up when an address was tested by
6191 -bt or -bv, which affected the outcome if any router or filter referred to
6194 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6195 log only when "message" or "log_message" are set. However, if one of the
6196 conditions was an address verification, or a call to a nested ACL, the
6197 messages generated by the underlying test were being passed through. This
6198 no longer happens. The underlying message is available in $acl_verify_
6199 message for both "message" and "log_message" expansions, so it can be
6200 passed through if needed.
6202 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6203 new expansion $bh_ to give the encoded byte string without charset
6204 translation. Translation happens only if iconv() is available; HAVE_ICONV
6205 indicates this at build time. HEADERS_CHARSET gives the charset to
6206 translate to; headers_charset can change it in the configuration, and
6207 "headers charset" can change it in an individual filter file.
6209 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6210 that creates RFC 2047 encoded "words" labels them as that charset instead
6211 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6212 expansion operator; (ii) when Exim creates a From: line for a local
6213 message; (iii) when a header line is rewritten to include a "phrase" part.
6215 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6216 buggy, causing it to skip the first lines of messages whose message ID
6217 ended in 'D'. This would not have bitten before Exim release 4.14, because
6218 message IDs were unlikely to end in 'D' before then. The effect was to have
6219 incorrect size information for certain domains.
6221 45. #include "config.h" was missing at the start of the crypt16.c module. This
6222 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6225 46. If there was a timeout during a "random" callout check, Exim treated it as
6226 a failure of the random address, and carried on sending RSET and the real
6227 address. If the delay was just some slowness somewhere, the response to the
6228 original RCPT would be taken as a response to RSET and so on, causing
6229 mayhem of various kinds.
6231 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6232 when I implemented it. It didn't allow for the fact that some option values
6233 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6234 the right test for positive values.
6236 48. Domain names in DNS records are case-independent. Exim always looks them up
6237 in lower case. Some resolvers return domain names in exactly the case they
6238 appear in the zone file, that is, they may contain uppercase letters. Not
6239 all resolvers do this - some return always lower case. Exim was treating a
6240 change of case by a resolver as a change of domain, similar to a widening
6241 of a domain abbreviation. This triggered its re-routing code and so it was
6242 trying to route what was effectively the same domain again. This normally
6243 caused routing to fail (because the router wouldn't handle the domain
6244 twice). Now Exim checks for this case specially, and just changes the
6245 casing of the domain that it ultimately uses when it transmits the message
6248 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6251 50. If a filter generated a file delivery with a non-absolute name (possible if
6252 no home directory exists for the router), the forbid_file option was not
6255 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6256 the existing equality matching.
6258 52. Exim was using ints instead of ino_t variables in some places where it was
6259 dealing with inode numbers.
6261 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6262 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6263 called TMPDIR, and if it finds it is different, it changes its value.
6265 54. The smtp_printf() function is now made available to local_scan() so
6266 additional output lines can be written before returning. There is also an
6267 smtp_fflush() function to enable the detection of a dropped connection.
6268 The variables smtp_input and smtp_batched_input are exported to
6271 55. Changed the default runtime configuration: the message "Unknown user"
6272 has been removed from the ACL, and instead placed on the localuser router,
6273 using the cannot_route_message feature. This means that any verification
6274 failures that generate their own messages won't get overridden. Similarly,
6275 the "Unrouteable address" message that was in the ACL for unverifiable
6276 relay addresses has also been removed.
6278 56. Added hosts_avoid_esmtp to the smtp transport.
6280 57. The exicyclog script was not checking for the esoteric option
6281 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6282 will work only if exicyclog is run under the appropriate euid.
6284 58. Following a discussion on the list, the rules by which Exim recognises line
6285 endings on incoming messages have been changed. The -dropcr and drop_cr
6286 options are now no-ops, retained only for backwards compatibility. The
6287 following line terminators are recognized: LF CRLF CR. However, special
6288 processing applies to CR:
6290 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6291 nor a local message in the state where . is a terminator.
6293 (ii) If a bare CR is encountered in a header line, an extra space is added
6294 after the line terminator so as not to end the header. The reasoning
6295 behind this is that bare CRs in header lines are most likely either
6296 to be mistakes, or people trying to play silly games.
6298 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6299 was being incorrectly given as 18 bytes larger than it should have been.
6300 This is a VOB (very old bug).
6302 60. This may never have affected anything current, but just in case it has:
6303 When the local host is found other than at the start of a list of hosts,
6304 the local host, those with the same MX, and any that follow, are discarded.
6305 When the list in question was part of a longer list of hosts, the following
6306 hosts (not currently being processed) were also being discarded. This no
6307 longer happens. I'm not sure if this situation could ever has previously
6310 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6313 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6314 header. This is something that is recommended in a new Internet Draft, and
6315 is something that is documented as being done by Sendmail. There are two
6316 possible values. For messages generated by the autoreply transport, Exim
6319 Auto-Submitted: auto-replied
6321 whereas for all other generated messages (e.g. bounces) it adds
6323 Auto-Submitted: auto-generated
6325 63. The "personal" condition in filters now includes a test for the
6326 Auto-Submitted: header. If it contains the string "auto-" the message it
6327 not considered personal.
6329 64. Added rcpt_include_affixes as a generic transport option.
6331 65. Added queue_only_override (default true).
6333 66. Added the syslog_duplication option.
6335 67. If what should have been the first header line of a message consisted of
6336 a space followed by a colon, Exim was mis-interpreting it as a header line.
6337 It isn't of course - it is syntactically invalid and should therefore be
6338 treated as the start of the message body. The misbehaviour could have
6339 caused a number of strange effects, including loss of data in subsequent
6340 header lines, and spool format errors.
6342 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6343 client host had authenticated. This control can now be exercised by an ACL
6344 for more flexibility.
6346 69. By default, callouts do not happen when testing with -bh. There is now a
6347 variant, -bhc, which does actually run the callout code, including
6348 consulting and updating the callout cache.
6350 70. Added support for saslauthd authentication, courtesy of Alexander
6353 71. If statvfs() failed on the spool or log directories while checking their
6354 size for availability, Exim confusingly gave the error "space shortage".
6355 Furthermore, in debugging mode it crashed with a floating point exception.
6356 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6357 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6358 serious problem, Exim now writes to the main and panic logs when this
6359 happens, with details of the failure. It then refuses to accept the
6360 incoming message, giving the message "spool directory problem" or "log
6361 directory problem" with a 421 code for SMTP messages.
6363 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6364 0, 1, and 2 exist, because some OS complain for execs without them (see
6365 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6366 descriptors. However, the code omitted to check that the open succeeded,
6367 causing mysterious errors if for some reason the permissions on /dev/null
6368 got screwed. Now Exim writes a message to the main and panic logs, and
6369 bombs out if it can't open /dev/null.
6371 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6372 interact so that it is all more flexible. It is supposed to remain
6373 backwards compatible. Also added extra_local_interfaces.
6375 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6376 to bomb out with an assertion failure - to the client this appears as a
6377 connection drop. This problem occurs in the part of the code that was taken
6378 from the Samba project. Fortunately, the assertion is in a very simple
6379 function, so I have fixed this by reproducing the function inline in the
6380 one place where it is called, and arranging for authentication to fail
6381 instead of killing the process with assert().
6383 75. The SPA client code was not working when the server requested OEM rather
6384 than Unicode encoding.
6386 76. Added code to make require_files with a specific uid setting more usable in
6387 the case where statting the file as root fails - usually a non-root-mounted
6388 NFS file system. When this happens and the failure is EACCES, Exim now
6389 forks a subprocess and does the per-uid checking as the relevant uid.
6391 77. Added process_log_path.
6393 78. If log_file_path was not explicitly set, a setting of check_log_space or
6394 check_log_inodes was ignored.
6396 79. If a space check for the spool or log partitions fails, the incident is now
6397 logged. Of course, in the latter case the data may get lost...
6399 80. Added the %p formatting code to string_format() so that it can be used to
6400 print addresses in debug_print(). Adjusted all the address printing in the
6401 debugging in store.c to use %p rather than %d.
6403 81. There was a concern that a line of code in smtp_in.c could overflow a
6404 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6405 initially expressed, the concern was not well-founded, because trailing
6406 spaces are removed early. However, if the trailing spaces were followed by
6407 a NULL, they did not get removed, so the overflow was possible. Two fixes
6410 (a) I re-wrote the offending code in a cleaner fashion.
6411 (b) If an incoming SMTP command contains a NULL character, it is rejected
6414 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6415 runs initgroups(), so that if the Exim user is in any additional groups,
6416 they will be used during message reception.
6422 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob